Daily Papers

For deploying foundation models, practitioners increasingly need prescriptive scaling laws: given a pre training compute budget, what downstream accuracy is attainable with contemporary post training practice, and how stable is that mapping as the field evolves? Using large scale observational evaluations with 5k observational and 2k newly sampled data on model performance, we estimate capability boundaries, high conditional quantiles of benchmark scores as a function of log pre training FLOPs, via smoothed quantile regression with a monotone, saturating sigmoid parameterization. We validate the temporal reliability by fitting on earlier model generations and evaluating on later releases. Across various tasks, the estimated boundaries are mostly stable, with the exception of math reasoning that exhibits a consistently advancing boundary over time. We then extend our approach to analyze task dependent saturation and to probe contamination related shifts on math reasoning tasks. Finally, we introduce an efficient algorithm that recovers near full data frontiers using roughly 20% of evaluation budget. Together, our work releases the Proteus 2k, the latest model performance evaluation dataset, and introduces a practical methodology for translating compute budgets into reliable performance expectations and for monitoring when capability boundaries shift across time.

Observational studies can yield clinically actionable evidence at scale, but executing them on real-world databases is open-ended and requires coherent decisions across cohort construction, analysis, and reporting. Prior evaluations of LLM agents emphasize isolated steps or single answers, missing the integrity and internal structure of the resulting evidence bundle. To address this gap, we introduce RWE-bench, a benchmark grounded in MIMIC-IV and derived from peer-reviewed observational studies. Each task provides the corresponding study protocol as the reference standard, requiring agents to execute experiments in a real database and iteratively generate tree-structured evidence bundles. We evaluate six LLMs (three open-source, three closed-source) under three agent scaffolds using both question-level correctness and end-to-end task metrics. Across 162 tasks, task success is low: the best agent reaches 39.9%, and the best open-source model reaches 30.4%. Agent scaffolds also matter substantially, causing over 30% variation in performance metrics. Furthermore, we implement an automated cohort evaluation method to rapidly localize errors and identify agent failure modes. Overall, the results highlight persistent limitations in agents' ability to produce end-to-end evidence bundles, and efficient validation remains an important direction for future work. Code and data are available at https://github.com/somewordstoolate/RWE-bench.

Benchmark suites assess model capability on controlled tasks; large-scale conversation corpora capture naturalistic use without user feedback; and in-interface feedback mechanisms record satisfaction without task purpose. Together, they leave a critical gap in LLM evaluation: no existing infrastructure routinely links interaction trajectories to user-defined outcomes. We introduce MonitrLLM, open-source infrastructure for community-centered LLM evaluations that links full conversation transcripts to user-reported task intent and outcome assessments, treating all three as primary evaluative signals rather than optional metadata. To demonstrate the value of this approach, we conducted a two-week feasibility pilot with 26 college students using ChatGPT, collecting 206 evaluation reports with full conversation transcripts. The findings from our pilot demonstrate the value of connecting conversation trajectories with user-reported outcomes. For instance, despite reporting high average satisfaction (4.19/5) with their LLM interactions, participants also experience a substantial 23.1% failure rate on their goal tasks. We also find that multi-turn conversations are reported as failing at 2.5 times the rate of single-turn exchanges, a pattern that reframes extended interaction as a signal of difficulty rather than engagement. We conclude by discussing the value of incorporating direct user feedback with observational data for robust LLM evaluations, and the possibilities for infrastructure that enables this goal.

Automated research systems support literature synthesis, ideation, experiments, writing, and peer review, but their evaluation is dispersed across tasks, benchmarks, and studies that are difficult to compare directly. We review this literature from the perspective of evaluation design and evidence, covering six targets: literature synthesis, research ideation, executable workflows, scholarly writing and communication, automatic peer review, and end-to-end research. We compare task construction, evidence sources, evaluators, and scoring procedures to explain the capabilities assessed by different designs. Our synthesis highlights three recurring lessons: output checks, process checks, and human studies provide complementary information; evaluator calibration is specific to the property being assessed; and resource budgets and attempt selection are integral to interpreting performance comparisons. We identify diagnostic evaluation designs and documented gaps in supporting evidence, and translate these comparisons into reporting and audit recommendations for specific evaluation settings. The survey helps readers navigate existing evaluations, select appropriate benchmarks, and design subsequent studies.

Observational studies require adjustment for confounding factors that are correlated with both the treatment and outcome. In the setting where the observed variables are tabular quantities such as average income in a neighborhood, tools have been developed for addressing such confounding. However, in many parts of the developing world, features about local communities may be scarce. In this context, satellite imagery can play an important role, serving as a proxy for the confounding variables otherwise unobserved. In this paper, we study confounder adjustment in this non-tabular setting, where patterns or objects found in satellite images contribute to the confounder bias. Using the evaluation of anti-poverty aid programs in Africa as our running example, we formalize the challenge of performing causal adjustment with such unstructured data -- what conditions are sufficient to identify causal effects, how to perform estimation, and how to quantify the ways in which certain aspects of the unstructured image object are most predictive of the treatment decision. Via simulation, we also explore the sensitivity of satellite image-based observational inference to image resolution and to misspecification of the image-associated confounder. Finally, we apply these tools in estimating the effect of anti-poverty interventions in African communities from satellite imagery.

Observational studies of causal effects require adjustment for confounding factors. In the tabular setting, where these factors are well-defined, separate random variables, the effect of confounding is well understood. However, in public policy, ecology, and in medicine, decisions are often made in non-tabular settings, informed by patterns or objects detected in images (e.g., maps, satellite or tomography imagery). Using such imagery for causal inference presents an opportunity because objects in the image may be related to the treatment and outcome of interest. In these cases, we rely on the images to adjust for confounding but observed data do not directly label the existence of the important objects. Motivated by real-world applications, we formalize this challenge, how it can be handled, and what conditions are sufficient to identify and estimate causal effects. We analyze finite-sample performance using simulation experiments, estimating effects using a propensity adjustment algorithm that employs a machine learning model to estimate the image confounding. Our experiments also examine sensitivity to misspecification of the image pattern mechanism. Finally, we use our methodology to estimate the effects of policy interventions on poverty in African communities from satellite imagery.

Evaluations are critical for understanding the capabilities of large language models (LLMs). Fundamentally, evaluations are experiments; but the literature on evaluations has largely ignored the literature from other sciences on experiment analysis and planning. This article shows researchers with some training in statistics how to think about and analyze data from language model evaluations. Conceptualizing evaluation questions as having been drawn from an unseen super-population, we present formulas for analyzing evaluation data, measuring differences between two models, and planning an evaluation experiment. We make a number of specific recommendations for running language model evaluations and reporting experiment results in a way that minimizes statistical noise and maximizes informativeness.

Benchmarks are necessary for healthcare evaluation, but are not sufficient for predicting deployment performance. Our position is that the evaluation--deployment gap arises not because of poorly designed benchmarks, but from implicit assumptions about how users interact with models that cannot be surfaced from benchmarks alone. To make this precise, we propose a classification of assumptions into two categories: task, which can be tested from conversation data alone, and outcome, which requires outcome data and behavioral studies for testing. Critically, outcome assumptions depend on human behavior, something that even well-designed benchmarks cannot directly observe. To demonstrate the operationality of this framework, we retrospectively analyze a healthcare RCT as a case study and find that the gap naturally separates into task and outcome gaps of roughly equal size. To address this, we make two contributions: first, we propose BenchmarkCards, an artifact that documents assumptions, and second, we propose staged evaluation, a procedure that systematically tests assumptions and evaluates performance.

Physical-field reconstruction and forecasting depend on both measurement density and spatial layout, yet evaluation under a single observation pattern does not characterize performance when that pattern changes. We introduce PDE-OBS, an integrated benchmarking platform spanning numerical data generation, model training, and inference and evaluation under varying observation conditions. It combines 560,000 fields and trajectories from seven partial differential equation families with configurable observation operators and seven adapted baseline methods for stationary reconstruction and short-horizon forecasting. Separating observation construction from physical records allows users to specify parameterized patterns and deterministic mixtures for training and testing while preserving prediction targets and data splits. The evaluation protocol uses references trained for each test pattern to compare models on identical test observations and targets, alongside equal-count groups for spatial-layout comparisons. On a 14,000-record subset, we evaluate 441 trained models under nine test patterns, yielding 3,969 evaluations. Mean cross-pattern error exceeds mean matched-pattern error in all 49 PDE-method pairs, and this finding persists in a configuration-matched subset of 117 models. Denser test observations do not consistently reduce error for a fixed model. Mixed-pattern training on five completed pairs reduces large single-pattern transfer errors, although destination-trained references usually remain more accurate. Together, the benchmark and findings support systematic evaluation of observation-pattern sensitivity and provide a reusable workflow for developing methods under changing measurement conditions. Code: https://github.com/ru1ch3n/PDE-OBS.

Language model users often issue queries that lack specification, where the context under which a query was issued -- such as the user's identity, the query's intent, and the criteria for a response to be useful -- is not explicit. For instance, a good response to a subjective query like "What book should I read next?" would depend on the user's preferences, and a good response to an open-ended query like "How do antibiotics work against bacteria?" would depend on the user's expertise. This makes evaluation of responses to such queries an ill-posed task, as evaluators may make arbitrary judgments about the response quality. To remedy this, we present contextualized evaluations, a protocol that synthetically constructs context surrounding an underspecified query and provides it during evaluation. We find that the presence of context can 1) alter conclusions drawn from evaluation, even flipping win rates between model pairs, 2) nudge evaluators to make fewer judgments based on surface-level criteria, like style, and 3) provide new insights about model behavior across diverse contexts. Specifically, our procedure uncovers an implicit bias towards WEIRD contexts in models' "default" responses and we find that models are not equally sensitive to following different contexts, even when they are provided in prompts.

The recent adoption of artificial intelligence in socio-technical systems raises concerns about the black-box nature of the resulting decisions in fields such as hiring, finance, admissions, etc. If data subjects -- such as job applicants, loan applicants, and students -- receive an unfavorable outcome, they may be interested in algorithmic recourse, which involves updating certain features to yield a more favorable result when re-evaluated by algorithmic decision-making. Unfortunately, when individuals do not fully understand the incremental steps needed to change their circumstances, they risk following misguided paths that can lead to significant, long-term adverse consequences. Existing recourse approaches focus exclusively on the final recourse goal but neglect the possible incremental steps to reach the goal with real-life constraints, user preferences, and model artifacts. To address this gap, we formulate a visual analytic workflow for incremental recourse planning in collaboration with AI/ML experts and contribute an interactive visualization interface that helps data subjects efficiently navigate the recourse alternatives and make an informed decision. We present a usage scenario and subjective feedback from observational studies with twelve graduate students using a real-world dataset, which demonstrates that our approach can be instrumental for data subjects in choosing a suitable recourse path.

Randomized controlled trials (RCTs) are the benchmark for causal inference, yet field implementation can deviate. We here present a remote audit - a design-based, preregistrable diagnostic that uses only pre-treatment satellite imagery to test whether assignment is independent of local conditions. The conditional randomization test of the remote audit evaluates whether treatment assignment is more predictable from pre-treatment satellite features than expected under the experiment's registered mechanism, providing a finite-sample valid, design-based diagnostic that requires no parametric assumptions. The procedure is finite-sample valid, honors blocks and clusters, and controls multiplicity across image models and resolutions via a max-statistic. We illustrate with two RCTs: Uganda's Youth Opportunities Program, where the audit corroborates randomization and flags selection and missing-data risks; and a school-based trial in Bangladesh, where assignment is highly predictable from pre-treatment features relative to the stated design, consistent with independent concerns about irregularities. Remote audits complement balance tests, lower early-stage costs, and enable rapid design checks when baseline surveys are expensive or infeasible.

Frontier language models sometimes recognize that they are being evaluated and adjust their behavior, undermining validity of benchmark results. Yet the field studies it without a shared foundation, conflating properties of the evaluation with properties of the model, and detection with behavioral response. We ground evaluation awareness in social psychology, decomposing it into an environment component (how recognizable the task is) and a model component that separates recognition from propensity to act on it. We operationalize the environment component through eight categorized trigger factors, such as placeholder entities and grading-style output formats, and study recognition and behavior through chain-of-thought monitoring. Across nine frontier models and four benchmarks, recognition rates depend on the specific pairing of model and benchmark rather than on either in isolation. Recognition rarely leads to behavioral change, and when it does, the direction depends on the type of evaluation perceived. Models are also more sensitive to safety than capability evaluations, placing safety benchmark validity at greater risk. To study which factors each model is sensitive to and how they interact, we propose EvalAwareBench, a factor-controlled benchmark of 100 paired safety-capability tasks where each of the eight factors can be independently toggled, varying evaluative signals while holding the underlying request fixed. Through EvalAwareBench, we find that no single factor uniformly affects all models, but stacking factors progressively raises evaluation awareness across all of them. Our framework and EvalAwareBench provide the tools to measure, attribute, and mitigate evaluation awareness, pointing to behavioral consistency under recognition as a promising path forward.

Representation engineering methods have recently shown promise for enabling efficient steering of model behavior. However, evaluation pipelines for these methods have primarily relied on subjective demonstrations, instead of quantitative, objective metrics. We aim to take a step towards addressing this issue by advocating for four properties missing from current evaluations: (i) contexts sufficiently similar to downstream tasks should be used for assessing intervention quality; (ii) model likelihoods should be accounted for; (iii) evaluations should allow for standardized comparisons across different target behaviors; and (iv) baseline comparisons should be offered. We introduce an evaluation pipeline grounded in these criteria, offering both a quantitative and visual analysis of how effectively a given method works. We use this pipeline to evaluate two representation engineering methods on how effectively they can steer behaviors such as truthfulness and corrigibility, finding that some interventions are less effective than previously reported.

Despite the success of reinforcement learning from human feedback (RLHF) in aligning language models, current reward modeling heavily relies on experimental feedback data collected from human annotators under controlled and costly conditions. In this work, we introduce observational reward modeling -- learning reward models with observational user feedback (e.g., clicks, copies, and upvotes) -- as a scalable and cost-effective alternative. We identify two fundamental challenges in this setting: (1) observational feedback is noisy due to annotation errors, which deviates it from true user preference; (2) observational feedback is biased by user preference, where users preferentially provide feedback on responses they feel strongly about, which creats a distribution shift between training and inference data. To address these challenges, we propose CausalRM, a causal-theoretic reward modeling framework that aims to learn unbiased reward models from observational feedback. To tackle challenge (1), CausalRM introduces a noise-aware surrogate loss term that is provably equivalent to the primal loss under noise-free conditions by explicitly modeling the annotation error generation process. To tackle challenge (2), CausalRM uses propensity scores -- the probability of a user providing feedback for a given response -- to reweight training samples, yielding a loss function that eliminates user preference bias. Extensive experiments across diverse LLM backbones and benchmark datasets validate that CausalRM effectively learns accurate reward signals from noisy and biased observational feedback and delivers substantial performance improvements on downstream RLHF tasks -- including a 49.2% gain on WildGuardMix and a 32.7% improvement on HarmBench. Code is available on our project website.

Most benchmarks for causal inference over time series are observational, small, or domain-specific, leaving interventional and counterfactual estimation under-served exactly where it matters most, such as in healthcare, policy evaluation, and climate science. We introduce DoTime, an open, scalable, and theoretically grounded generator of multivariate temporal structural causal models (TSCMs) with interventions, released as the dotime PyPI package together with four frozen evaluation suites. Beyond existing work, it adds capabilities absent from prior generators: continuous-time intervention windows, counterfactual sampling modes with a positivity guard, regime-switching SCMs as a strict generalization of interrupted time series, non-stationary dynamics by construction with switching SCM parameters, and deterministic ramp and sinusoidal intervention profiles that place trends and structural breaks inside the evaluation window. Moreover, it demonstrates the suitability of the generator as a prior for a causal foundation model reference implementation. The released suites span a training-scale snapshot of 100{,}000 trajectories and eight named identification structures, each with exact ground truth: paired interventional trajectories from the same SCM throughout, and shared-noise counterfactuals in the continuous-time suite. We ship reference baseline implementations with an evaluation harness, and pose a falsifiable claim: interventional training buys a measurable direction-accuracy advantage over an observational model of identical capacity. It is tested across three training seeds per arm. Under structure-matched evaluation on held-out episodes, the interventional prior-fitted network's (PFN) gap is positive in every structure, trajectory length, and seed tested.

Recent advances have made long-form report-generating systems widely available. This has prompted evaluation frameworks that use LLM-as-judge protocols and claim verification, along with meta-evaluation frameworks that seek to validate these methods. Many of the meta-evaluations estimate an evaluation quality's by comparing its assessments against human pairwise preferences. Prior work, however, suggests that human pairwise preference may be overly simplistic and can fail to capture nuances of expert expectations. We conduct a case study in meta-evaluation for long-form QA benchmarks using ScholarQA-CS2, a benchmark designed for assessing retrieval-augmented deep-research QA in the scientific domain. We comprehensively validate the benchmark through human pairwise preference judgments, then critically examine the strengths, weaknesses, and confounders of this approach. We show that pairwise preference rankings are best suited for system-level evaluation, while explicit metric-wise annotations and expert annotators are critical for reliable metric-level assessment, with subjectivity remaining a key challenge. Based on our findings, we offer practical guidelines for designing future meta-evaluations that better align evaluation methods, annotator expertise, and reporting practices. By surfacing these methodological challenges, we aim to advance evaluation standards for deep-research systems.

Debates about whether development projects improve living conditions persist, partly because observational estimates can be biased by incomplete adjustment and because reliable outcome data are scarce at the neighborhood level. We address both issues in a continent-scale, sector-specific evaluation of Chinese and World Bank projects across 9,899 neighborhoods in 36 African countries (2002 to 2013), representative of 88% of the population. First, we use a recent dataset that measures living conditions with a machine-learned wealth index derived from contemporaneous satellite imagery, yielding a consistent panel of 6.7 km square mosaics. Second, to strengthen identification, we proxy officials' map-based placement criteria using pre-treatment daytime satellite images and fuse these with rich tabular covariates to estimate funder- and sector-specific ATEs via inverse-probability weighting. Incorporating imagery systematically shrinks effects relative to tabular-only models, indicating prior work likely overstated benefits. On average, both donors raise wealth, with larger gains for China; sector extremes in our sample include Trade and Tourism for the World Bank (+6.27 IWI points), and Emergency Response for China (+14.32). Assignment-mechanism analyses show World Bank placement is generally more predictable from imagery alone, as well as from tabular covariates. This suggests that Chinese project placements are more driven by non-visible, political, or event-driven factors than World Bank placements. To probe residual concerns about selection on observables, we also estimate within-neighborhood (unit) fixed-effects models at a spatial resolution about 450 times finer than prior fixed effects analyses, leveraging the computer-vision-imputed IWI panels; these deliver smaller but directionally consistent effects.

Large language models are being widely used across industries to generate content that contributes directly to key performance metrics, such as conversion rates. Pretrained models, however, often fall short when it comes to aligning with human preferences or optimizing for business objectives. As a result, fine-tuning with good-quality labeled data is essential to guide models to generate content that achieves better results. Controlled experiments, like A/B tests, can provide such data, but they are often expensive and come with significant engineering and logistical challenges. Meanwhile, companies have access to a vast amount of historical (observational) data that remains underutilized. In this work, we study the challenges and opportunities of fine-tuning LLMs using observational data. We show that while observational outcomes can provide valuable supervision, directly fine-tuning models on such data can lead them to learn spurious correlations. We present empirical evidence of this issue using various real-world datasets and propose DeconfoundLM, a method that explicitly removes the effect of known confounders from reward signals. Using simulation experiments, we demonstrate that DeconfoundLM improves the recovery of causal relationships and mitigates failure modes found in fine-tuning methods that ignore or naively incorporate confounding variables. Our findings highlight that while observational data presents risks, with the right causal corrections, it can be a powerful source of signal for LLM alignment. Please refer to the project page for code and related resources.

We develop and evaluate a data-driven approach for detecting unusual (anomalous) patient-management actions using past patient cases stored in an electronic health record (EHR) system. Our hypothesis is that patient-management actions that are unusual with respect to past patients may be due to a potential error and that it is worthwhile to raise an alert if such a condition is encountered. We evaluate this hypothesis using data obtained from the electronic health records of 4,486 post-cardiac surgical patients. We base the evaluation on the opinions of a panel of experts. The results support that anomaly-based alerting can have reasonably low false alert rates and that stronger anomalies are correlated with higher alert rates.

The increasing threat of emerging infectious diseases and antimicrobial resistance requires more efficient, high-quality research. Perpetual Observational Studies (POS) nested within a clinical research network can improve planning, quality and efficiency of interventional and observational studies, although real-life benefits and challenges need to be assessed. Ecraid (European Clinical Research Alliance on Infectious Diseases) has initiated POS and will monitor the impact for five specific infectious syndromes.

Longitudinal observational data are often subject to informative missingness. For example, in healthcare research using electronic health records, covariates and outcomes are measured only when patients interact with the healthcare system. Whether these data are observed usually depends on the unobserved variable values themselves, introducing a common form of data missing not at random. We develop identification and estimation methods for the effects of longitudinal modified treatment policies under a self-censoring model which explicitly models this dependence. We consider effects of treatments that can be modified only at measurement times (e.g., interactions with the heathcare system). We study two settings: one in which the observed history suffices for treatment confounding control, and another in which confounder control requires covariate values that were not measured at missed visits. In both settings, we combine longitudinal causal identification with shadow-variable conditional moment restrictions to obtain recursive identification formulas for the counterfactual outcome mean. We derive gradients of the pathwise derivatives and second-order remainder expansions, and use these results to construct estimators that accommodate machine learning for nuisance-function estimation. Under suitable product-rate conditions and cross-fitting, the estimators are asymptotically linear and permit root-n inference, including simultaneous inference across outcome times. We also describe conditional-moment learning and ensemble approaches for estimating the nuisance functions required by our approach, together with a pool-and-smooth strategy that borrows information across sparsely measured outcomes.

Both capability and safety benchmarks rest upon the assumption that the behavior of language models undergoing a test is informative about their behavior in deployment. This assumption can fail, should models infer that they are being evaluated and condition their response on such context. This hypothesis, termed ``evaluation awareness'', has been observed in frontier and open-weight language models alike. We provide a systematic study of this phenomenon, by probing for it across six language models (from four families and three sizes) and three metrics. More precisely, we examine whether (i) being under evaluation is linearly represented within the models' activations space, (ii) it is verbalized in their output tokens (as scored by an LLM-as-judge), and (iii) steering causally affects their behavior. For the open-checkpoint Olmo models, we further test these measures at every training stage. In doing so, we report that evaluation awareness is linearly decodable from the residual streams of every model (best AUROC geq 0.7). By contrast, these representations align only in part with verbalization: their correlations and mutual information are nonzero in some settings, yet vary substantially across models, layers, and readout choices. Nevertheless, steering along probe-derived directions can shift the verbalization scores. Finally, a comparison across the Olmo checkpoints reveals that evaluation awareness is already present within base models, becomes amplified throughout the stages of supervised fine-tuning, and remains stable thereafter---unlike the effects of steering, that grow more pronounced at every successive training stage. These results show the need for evaluations to account for the disjunction between what models represent internally, what they verbalize, and their steering.

Data-driven social science research is inherently slow, relying on iterative cycles of observation, hypothesis generation, and experimental validation. While recent data-driven methods promise to accelerate parts of this process, they largely fail to support end-to-end scientific discovery. To address this gap, we introduce EXPERIGEN, an agentic framework that operationalizes end-to-end discovery through a Bayesian optimization inspired two-phase search, in which a Generator proposes candidate hypotheses and an Experimenter evaluates them empirically. Across multiple domains, EXPERIGEN consistently discovers 2-4x more statistically significant hypotheses that are 7-17 percent more predictive than prior approaches, and naturally extends to complex data regimes including multimodal and relational datasets. Beyond statistical performance, hypotheses must be novel, empirically grounded, and actionable to drive real scientific progress. To evaluate these qualities, we conduct an expert review of machine-generated hypotheses, collecting feedback from senior faculty. Among 25 reviewed hypotheses, 88 percent were rated moderately or strongly novel, 70 percent were deemed impactful and worth pursuing, and most demonstrated rigor comparable to senior graduate-level research. Finally, recognizing that ultimate validation requires real-world evidence, we conduct the first A/B test of LLM-generated hypotheses, observing statistically significant results with p less than 1e-6 and a large effect size of 344 percent.

Difference-in-differences (DID) studies are widely used to evaluate climate policy, but assessing the evidence supporting their identification assumptions remains challenging. We introduce ARGUS, a structured language-model pipeline that audits reported evidence against an eleven-dimension assumption-implication-evidence rubric and abstains when relevant evidence cannot be retrieved. We evaluate ARGUS using injected flaws, economics papers, and a small pilot with reconciled labels. On the 11-flaw benchmark, ARGUS detects 73% of planted flaws, compared with 18% for a keyword-based pipeline. Across 26 economics papers, ARGUS abstains on about 40% of paper-dimension assessments for lack of retrievable evidence. In a five-paper pilot with labels reconciled by two annotators, it assigns a higher risk level than the labels on 25 of the 33 assessments it completes. A rule fixed before the labels arrived removes most of this in-sample; weighted agreement stays low. ARGUS provides evidence-linked risk reports that localize potential weaknesses for expert review, without adjudicating causal claims. Code and data: https://github.com/yonghongzhang-io/ARGUS

Evaluations (Evals) are a deployment bottleneck for real-world AI applications: public benchmarks rarely match a team's users, context, or policies, and human review is often tedious to scale. Motivated by our work with AI applications in the public sector, this project addresses recurring evaluation challenges encountered when applications must satisfy local policy and governance requirements. We present Kaleidoscope, an integrated workflow for contextual functional evaluation that links persona-based test generation, contextualized rubrics, and human review for reliability-gated automated scoring. Generated test cases are scored against application-specific rubrics; human annotations provide reviewable labels; and LLM judges automate scoring only when their agreement with those labels meets a configured threshold. Kaleidoscope is therefore a practical, inspectable, iterative workflow for product teams. We report early evidence from a three-week pilot across four organizational use cases and custom-rubric judge experiments on 108 annotated Q\&A pairs spanning four domains and 14 evaluation dimensions. The results highlight useful features for end-to-end reliable, automated scoring.

The "thinking-with-images" paradigm equips multimodal LLMs with active visual operations such as crop-and-zoom. However, models using these operations often achieve only marginal or negative gains over direct inference at substantially higher token cost. They may also repeatedly crop irrelevant regions and fail on questions that direct inference answers correctly. We ask whether the returned visual evidence causally affects the answer. To answer this question, we formulate visual tool-use as a causal graph that separates observation-mediated paths from action-induced shortcuts. We then audit it through interventions at the three levels: policy (comparing tool-use with direct inference), trajectory (corrupting all observations during rollout), and step (counterfactually replacing one individual observation under a fixed prefix). Our step-level estimand, Visual Evidence Gain, isolates the contribution of each returned observation. Across six representative models and five fine-grained perception benchmarks, we uncover policy miscalibration with two failure modes. In Calling Without Looking, returned observations have no causal effect on the answer. In Looking Without Planning, observations are informative but the call schedule is incoherent. A trajectory-level diagnostic decomposes the policy-level accuracy gain and shows that the gain is concentrated in a Calibrated minority. We term this discrepancy the illusion of visual tool-use: despite aggregate accuracy gains, visual tool-use is not causally effective across a broad range of rollouts. The code is available at https://github.com/OpenCausaLab/CauAudit.

Recent research in the field of reflexive metrics have studied the emergence and consequences of evaluation gaps in science. The concept of evaluation gaps captures potential discrepancies between what researchers value about their research, in particular research quality, and what metrics measure. As a result, scientists may experience anomie and adopt innovative ways to cope. These often value quantity over quality and may even compromise research integrity. A consequence of such gaps may therefore be research misconduct and a decrease in research quality. In the language of rational choice theory, an evaluation gap persists if motivational factors arising out of the internal component of an actors situation are incongruent with those arising out of the external components. The aim of this research is therefore to study and compare autonomous and controlled motivations to become an astronomer, to do research in astronomy and to publish scientific papers. Moreover, we study how these different motivational factors affect publication pressure, the experience of organisational justice and the observation of research misconduct. In summary, we find evidence for an evaluation gap and that controlled motivational factors arising from evaluation procedures based on publication record drives up publication pressure, which, in turn, was found to increase the likelihood of perceived frequency of misbehaviour.

In machine learning systems, bias mitigation approaches aim to make outcomes fairer across privileged and unprivileged groups. Bias mitigation methods work in different ways and have known "waterfall" effects, e.g., mitigating bias at one place may manifest bias elsewhere. In this paper, we aim to characterise impacted cohorts when mitigation interventions are applied. To do so, we treat intervention effects as a classification task and learn an explainable meta-classifier to identify cohorts that have altered outcomes. We examine a range of bias mitigation strategies that work at various stages of the model life cycle. We empirically demonstrate that our meta-classifier is able to uncover impacted cohorts. Further, we show that all tested mitigation strategies negatively impact a non-trivial fraction of cases, i.e., people who receive unfavourable outcomes solely on account of mitigation efforts. This is despite improvement in fairness metrics. We use these results as a basis to argue for more careful audits of static mitigation interventions that go beyond aggregate metrics.

In psychotherapy, therapeutic outcome assessment, or treatment outcome evaluation, is essential for enhancing mental health care by systematically evaluating therapeutic processes and outcomes. Existing large language model approaches often focus on therapist-centered, single-session evaluations, neglecting the client's subjective experience and longitudinal progress across multiple sessions. To address these limitations, we propose IPAEval, a client-Informed Psychological Assessment-based Evaluation framework that automates treatment outcome evaluations from the client's perspective using clinical interviews. IPAEval integrates cross-session client-contextual assessment and session-focused client-dynamics assessment to provide a comprehensive understanding of therapeutic progress. Experiments on our newly developed TheraPhase dataset demonstrate that IPAEval effectively tracks symptom severity and treatment outcomes over multiple sessions, outperforming previous single-session models and validating the benefits of items-aware reasoning mechanisms.

As the use of text-to-image generative models increases, so does the adoption of automatic benchmarking methods used in their evaluation. However, while metrics and datasets abound, there are few unified benchmarking libraries that provide a framework for performing evaluations across many datasets and metrics. Furthermore, the rapid introduction of increasingly robust benchmarking methods requires that evaluation libraries remain flexible to new datasets and metrics. Finally, there remains a gap in synthesizing evaluations in order to deliver actionable takeaways about model performance. To enable unified, flexible, and actionable evaluations, we introduce EvalGIM (pronounced ''EvalGym''), a library for evaluating generative image models. EvalGIM contains broad support for datasets and metrics used to measure quality, diversity, and consistency of text-to-image generative models. In addition, EvalGIM is designed with flexibility for user customization as a top priority and contains a structure that allows plug-and-play additions of new datasets and metrics. To enable actionable evaluation insights, we introduce ''Evaluation Exercises'' that highlight takeaways for specific evaluation questions. The Evaluation Exercises contain easy-to-use and reproducible implementations of two state-of-the-art evaluation methods of text-to-image generative models: consistency-diversity-realism Pareto Fronts and disaggregated measurements of performance disparities across groups. EvalGIM also contains Evaluation Exercises that introduce two new analysis methods for text-to-image generative models: robustness analyses of model rankings and balanced evaluations across different prompt styles. We encourage text-to-image model exploration with EvalGIM and invite contributions at https://github.com/facebookresearch/EvalGIM/.

Benchmarks can run without determining what their results license. We freeze a large evaluation collection and attempt to replay its historical claims. Most units stop because the evidence required for replay is not bound. Where replay is possible, different claims remain stable at different resolutions. We make this otherwise implicit inference step explicit and executable.

Helpful reviews have been essential for the success of e-commerce services, as they help customers make quick purchase decisions and benefit the merchants in their sales. While many reviews are informative, others provide little value and may contain spam, excessive appraisal, or unexpected biases. With the large volume of reviews and their uneven quality, the problem of detecting helpful reviews has drawn much attention lately. Existing methods for identifying helpful reviews primarily focus on review text and ignore the two key factors of (1) who post the reviews and (2) when the reviews are posted. Moreover, the helpfulness votes suffer from scarcity for less popular products and recently submitted (a.k.a., cold-start) reviews. To address these challenges, we introduce a dataset and develop a model that integrates the reviewer's expertise, derived from the past review history of the reviewers, and the temporal dynamics of the reviews to automatically assess review helpfulness. We conduct experiments on our dataset to demonstrate the effectiveness of incorporating these factors and report improved results compared to several well-established baselines.

AI evaluation results are produced at scale but reported inconsistently across leaderboards, model cards, benchmark papers, and company blogs. The cost is interpretive: readers cannot reliably compare results across sources, identify what a report omits, or trace an aggregate claim to its underlying evidence. Recent efforts address isolated components but leave three gaps: they cover only narrow slices of the evaluation lifecycle and do not compose into a single interpretable record; they specify static representations that do not differentiate the questions different stakeholders bring to the same evidence; and they remain proposals on paper, lacking the extraction infrastructure required for adoption at scale. We present , an operational reporting layer that composes benchmark metadata, evaluation run data, and model metadata into a unified record. We (1) derive a reporting schema from a structured review of 52 papers and 10 stakeholder interviews, (2) implement four interpretive signals (reproducibility, documentation completeness, provenance and risk, and score comparability), rendered through reader modes calibrated to research and non-research audiences, and (3) deploy a monitoring tool that applies across 5,816 models, 635 benchmarks, and 101,843 results, surfacing systematic gaps in current reporting practice.

The validity of AI safety evaluations depends on models behaving consistently across controlled and deployment settings. Prior work has identified test-time contextual cues, such as hypothetical scenarios, as a source of verbalized evaluation awareness and subsequent behavioral shift. In this paper, we investigate a potential explanation of this phenomenon: evaluation meta-knowledge, defined as parametric knowledge about the structural traits that characterize evaluations. Similar to dataset contamination, where benchmark exposure leads to higher performance through memorization, we hypothesize that models trained on texts describing evaluation practices may implicitly learn to recognize and respond to evaluation-like contexts, for instance, through exposure to scientific articles or social media posts about AI benchmarking. To test this, we fine-tune models on synthetic documents describing evaluation traits such as verifiable structures or moral dilemmas. Evaluating this fine-tuned model on six safety benchmarks, we find that it is significantly safer than the base model and control model. This behavioral shift persists even when restricting the analysis to responses lacking explicit verbalization of evaluation awareness. Our results demonstrate that evaluation meta-knowledge may inflate safety benchmark performance, introducing a novel confounder that is independent of explicit memorization or verbalized evaluation awareness, thus, challenging to detect. These findings have important implications for the design and interpretation of AI safety evaluations. Our code and models are available at https://github.com/compass-group-tue/arxiv2026_evaluation_meta_knowledge.

Deep research agents increasingly automate complex information-seeking tasks, producing evidence-grounded reports via multi-step reasoning, tool use, and synthesis. Their growing role demands scalable, reliable evaluation, positioning LLM-as-judge as a supervision paradigm for assessing factual accuracy, evidence use, and reasoning quality. Yet the reliability of these judges for deep research agents remains poorly understood, posing a critical meta-evaluation problem: before deploying LLM judges to supervise research agents, we must first evaluate the judges themselves. Existing meta-evaluations fall short in two ways: (1) reliance on coarse, subjective human-preference agreement; (2) focus on instruction-following or verifiable tasks, leaving open-ended agent executions unexplored. To address these gaps, we introduce REFLECT (REliable Fine-grained LLM judge Evaluation via Controlled inTervention), a meta-evaluation benchmark targeting fine-grained failure detection in agentic environments. REFLECT defines a detailed taxonomy of process- and outcome-level failure modes, instantiated by performing controlled and localized interventions on quality-screened agent execution traces. This yields verifiable, comprehensive, and fine-grained instances for validating the judge models. Our experiments show that current LLM judges remain unreliable: even the best-performing models achieve overall accuracies below 55% across reasoning, tool-use, and report-quality failures, with especially poor performance on evidence verification. Together, our taxonomy and findings expose systematic judge limitations, reveal tradeoffs in cost and reliability, and offer actionable guidance for building more reliable evaluation pipelines for deep research agents.

Clinical case formulation organizes patient symptoms and psychosocial factors into causal models, often using the 5P framework. However, constructing such graphs from therapy transcripts is time consuming and varies across clinicians. We present InsightFlow, an LLM based approach that automatically generates 5P aligned causal graphs from patient-therapist dialogues. Using 46 psychotherapy intake transcripts annotated by clinical experts, we evaluate LLM generated graphs against human formulations using structural (NetSimile), semantic (embedding similarity), and expert rated clinical criteria. The generated graphs show structural similarity comparable to inter annotator agreement and high semantic alignment with human graphs. Expert evaluations rate the outputs as moderately complete, consistent, and clinically useful. While LLM graphs tend to form more interconnected structures compared to the chain like patterns of human graphs, overall complexity and content coverage are similar. These results suggest that LLMs can produce clinically meaningful case formulation graphs within the natural variability of expert practice. InsightFlow highlights the potential of automated causal modeling to augment clinical workflows, with future work needed to improve temporal reasoning and reduce redundancy.

Foundation models are increasingly used in scientific research, but evaluating AI-generated scientific work remains challenging. While expert reviews are costly, large language models (LLMs) as proxy reviewers have proven to be unreliable. To address this, we investigate two automatic evaluation metrics, specifically citation count prediction and review score prediction. We parse all papers of OpenReview and augment each submission with its citation count, reference, and research hypothesis. Our findings reveal that citation count prediction is more viable than review score prediction, and predicting scores is more difficult purely from the research hypothesis than from the full paper. Furthermore, we show that a simple prediction model based solely on title and abstract outperforms LLM-based reviewers, though it still falls short of human-level consistency.

Reality may establish that an outcome occurred without identifying which evolving procedure produced it or why. This distinction matters in production ML systems whose code, configuration, and artifacts change while external feedback accumulates. We examine it in a human-directed, agent-engineered quantitative trading system, using oracle to mean an external source of realized outcomes rather than a complete correctness specification. Across one year, the account gained and outperformed a broad market index, while annual alpha was not statistically distinguishable from zero under the main retrospective specification. Retrospectively selected subperiods include adverse relative performance and conditional candidate-level weakness under declared approximate references. Engineering records document changes during the episode, and complete recommendation-to-runtime binding is unavailable. The archive does not establish a common frozen instance or a unique cause. The case motivates an outcome--diagnosis gap: outcome evidence, evaluated-object identity, and causal explanation support distinct claims. We distinguish frozen instances, pre-specified adaptive procedures, and ad-hoc development; organize archive-relative claim identifiability and an evidence hierarchy; and propose a prospective production-binding protocol. An illustrative compatible-history example shows how factual binding can resolve a recommendation's referent without supplying its counterfactual effect. The protocol is proposed rather than prospectively validated. External feedback constrains outcome claims, while provenance and additional identification structure determine the resolution of diagnosis.

Great scientists have strong judgement and foresight, closely tied to what we call scientific taste. Here, we use the term to refer to the capacity to judge and propose research ideas with high potential impact. However, most relative research focuses on improving an AI scientist's executive capability, while enhancing an AI's scientific taste remains underexplored. In this work, we propose Reinforcement Learning from Community Feedback (RLCF), a training paradigm that uses large-scale community signals as supervision, and formulate scientific taste learning as a preference modeling and alignment problem. For preference modeling, we train Scientific Judge on 700K field- and time-matched pairs of high- vs. low-citation papers to judge ideas. For preference alignment, using Scientific Judge as a reward model, we train a policy model, Scientific Thinker, to propose research ideas with high potential impact. Experiments show Scientific Judge outperforms SOTA LLMs (e.g., GPT-5.2, Gemini 3 Pro) and generalizes to future-year test, unseen fields, and peer-review preference. Furthermore, Scientific Thinker proposes research ideas with higher potential impact than baselines. Our findings show that AI can learn scientific taste, marking a key step toward reaching human-level AI scientists.

Agents can turn shared infrastructure into a channel for coordinated intrusion. The Hugging Face incident and a separate public-wiki investigation show why a security assessment may need evidence from several executions and the artifacts they leave behind. We argue that the operational unit of defence should be a revisable coordination episode linking observed transfers, task authority, and response history. The central research problem is prospective episode discovery: finding which actions belong together before an evaluator supplies their membership. We define unsanctioned coordination relative to collaboration and delegated-authority policy, connect storage-mediated coordination to stigmergy, and specify the evidence needed to distinguish influence from common causes. First-contact signals are one possible input to discovery; the design also follows inherited state and later use. A proposed evaluation compares isolated actions, rolling windows, known groups, and prospectively discovered episodes at matched review cost and false-alert workload. It measures harmful outcomes across all assigned population runs and tests recurrence after channel closure and state quarantine. A checksum-verified reconstruction of the public wiki export separates the decline in retained writes from later administrative cleanup. The contribution is an incident-grounded position, descriptive analysis, and evaluation design. It makes the recommendation to monitor across executions testable without claiming a new detector or a measured containment benefit.

Safety benchmarks assume that test-condition behavior predicts deployment behavior, an assumption that fails if models detect evaluation cues and adapt. This opens a gap between benchmark performance and deployment behavior: compliance measured under test conditions becomes an optimistic upper bound that overstates how safely a model behaves once the evaluation harness is removed. We characterize this evaluation awareness through eight experiments across 37 open-weight models and seven families. (i)Detection is moderate and training-driven (24/37 models exceed chance, best AUROC 0.714 vs.0.819 human, with instruction tuning dominating over scale). (ii)Detection shifts safety behavior (hard refusal drops 5.8 percentage points under hypothetical framing, and 21/140 HarmBench framing effects are significant, with compliance rising up to +30 percentage points. (iii)Representations survive behavioral collapse (probes retain AUROC 0.98 under rewrites that drive behavior below chance, and multi-layer steering causally moves three downstream tasks while random controls do not). (iv)These axes are weakly coupled (only 1/15 correlations are significant, the sole robust link being behavioral detection versus framing resistance, ρ=-0.79, p<0.001). We call this gap the benchmark illusion: because detectability, behavioral manifestation, and controllability vary independently, it is multivariate rather than a single number, so no single awareness score is a reliable proxy for deployment safety.

AI evaluation is undergoing a structural change. Large language models (LLMs) are increasingly deployed as systems that act over time through tools, environments, users, and other agents, while many evaluation practices still inherit assumptions from response-centered benchmarks (e.g., fixed inputs, isolated outputs, and outcome judgments that can be made from a single response). The field has begun to build interactive benchmarks, but the resulting landscape is fragmented: benchmarks differ in what interaction artifacts they admit, how trajectories are scored, and what claims their results support. This position paper argues that interactive evaluation should be treated as a principled evaluation paradigm, not merely a new family of agent benchmarks. Simply adopting previous evaluation paradigms does not suffice. We define evaluation as an autonomous mapping from evidence to judgments, and show that interactive evaluation changes both sides of this mapping: the evidence becomes interaction-generated trajectories, while the evaluation procedure must assess process, recoverability, coordination, robustness, and system-level performance. Building on this definition, we propose a two-axis taxonomy, derive design principles and reporting standards, examine representative scenarios, and analyze how longstanding evaluation challenges reappear at the trajectory level.

This paper presents an eight-week observational comparison of 68 single-file HTML generations collected across 17 public experiments in the "HTML AI Battle" project between December 10, 2025 and February 4, 2026. Four reasoning model families, GPT, Gemini, Grok, and Claude, were compared under a fixed public-interface protocol with no custom instructions, no personality tuning, and no repair prompts. Each output was evaluated from a rendered browser video using human scores and a Gemini LLM-as-a-judge layer for prompt adherence, functional correctness, and UI quality, then packaged into a standardized social-media protocol spanning X (Twitter), TikTok, and YouTube. The tracker was also used for two supervised predictive analyses: an experiment-level model for 24-hour X impressions and a generation-level model for HTML verbosity. Under this protocol, Claude was the strongest and most consistent family, leading mean performance and winning 9/17 prompts under the primary human weighted score. Longer measured reasoning time was not associated with higher quality overall. Gemini as a judge was significantly more lenient than the human evaluator on functional correctness and overall performance, while stable self-favoring bias remained unresolved. The exploratory X-impressions model remained weak under post-screen cross-validation (MAE = 46,874, R^2 = -0.377), whereas the HTML-lines model performed better, with a model-family-only baseline outperforming prompt-aware alternatives (MAE = 135.2, R^2 = 0.576). Overall, selected pre-publication technical/audio variables were not sufficient to predict 24-hour X reach, while code verbosity was driven much more by model family than by prompt wording. The comparisons remain observational and are limited by public-interface drift, access-path differences, and one primary human scorer.

Evaluation has traditionally focused on ranking candidates for a specific skill. Modern generalist models, such as Large Language Models (LLMs), decidedly outpace this paradigm. Open-ended evaluation systems, where candidate models are compared on user-submitted prompts, have emerged as a popular solution. Despite their many advantages, we show that the current Elo-based rating systems can be susceptible to and even reinforce biases in data, intentional or accidental, due to their sensitivity to redundancies. To address this issue, we propose evaluation as a 3-player game, and introduce novel game-theoretic solution concepts to ensure robustness to redundancy. We show that our method leads to intuitive ratings and provide insights into the competitive landscape of LLM development.

Red-team evaluations of AI models support some claims and not others, and the boundary between the two is calculable rather than merely a matter of judgment. We define the evidential ceiling of an evaluation as the largest factor by which one result can move belief under a fixed testing budget, derive it in closed form for the benchmark null result, and use it to locate that boundary exactly. We find that above a calculable harm rate, a benchmark of modest size certifies a category to a stated evidentiary standard, and a clean sheet is then the stronger of the two possible observations, outweighing a single reproduced failure. Below that rate, no passive benchmark of feasible size provides the specified evidence of safety under the fixed scoring rule and approximately independent trial structure. The crossing between the two regimes has a closed form. The bound is not specific to benchmarks: written in terms of a procedure's hypothesis conditioned elicitation rates, it covers adaptive and automated red teaming as well, and shows that discrimination between the hypotheses rather than attack success is what determines evidential worth. Auditing eight evaluation suites against the boundary, we find that current benchmarks are adequate for high-frequency harm categories and several orders of magnitude short for rare, catastrophic ones. Safety benchmarks are not uninformative. They are informative about a specific and computable set of propositions, and the discipline they need is to state which.

We develop and evaluate a data-driven approach for detecting unusual (anomalous) patient-management decisions using past patient cases stored in electronic health records (EHRs). Our hypothesis is that a patient-management decision that is unusual with respect to past patient care may be due to an error and that it is worthwhile to generate an alert if such a decision is encountered. We evaluate this hypothesis using data obtained from EHRs of 4486 post-cardiac surgical patients and a subset of 222 alerts generated from the data. We base the evaluation on the opinions of a panel of experts. The results of the study support our hypothesis that the outlier-based alerting can lead to promising true alert rates. We observed true alert rates that ranged from 25\% to 66\% for a variety of patient-management actions, with 66\% corresponding to the strongest outliers.

We present an automated, contrastive evaluation pipeline for auditing the behavioral impact of interventions on large language models. Given a base model M_1 and an intervention model M_2, our method compares their free-form, multi-token generations across aligned prompt contexts and produces human-readable, statistically validated natural-language hypotheses describing how the models differ, along with recurring themes that summarize patterns across validated hypotheses. We evaluate the approach in synthetic setting by injecting known behavioral changes and showing that the pipeline reliably recovers them. We then apply it to three real-world interventions, reasoning distillation, knowledge editing and unlearning, demonstrating that the method surfaces both intended and unexpected behavioral shifts, distinguishes large from subtle interventions, and does not hallucinate differences when effects are absent or misaligned with the prompt bank. Overall, the pipeline provides a statistically grounded and interpretable tool for post-hoc auditing of intervention-induced changes in model behavior.

This paper introduces LegalRikai: Open Benchmark, a new benchmark comprising four complex tasks that emulate Japanese corporate legal practices. The benchmark was created by legal professionals under the supervision of an attorney. This benchmark has 100 samples that require long-form, structured outputs, and we evaluated them against multiple practical criteria. We conducted both human and automated evaluations using leading LLMs, including GPT-5, Gemini 2.5 Pro, and Claude Opus 4.1. Our human evaluation revealed that abstract instructions prompted unnecessary modifications, highlighting model weaknesses in document-level editing that were missed by conventional short-text tasks. Furthermore, our analysis reveals that automated evaluation aligns well with human judgment on criteria with clear linguistic grounding, and assessing structural consistency remains a challenge. The result demonstrates the utility of automated evaluation as a screening tool when expert availability is limited. We propose a dataset evaluation framework to promote more practice-oriented research in the legal domain.

Psychotherapy or counseling is typically conducted through spoken conversation between a therapist and a client. Analyzing the speech characteristics of psychotherapeutic interactions can help understand the factors associated with effective psychotherapy. This paper introduces CUEMPATHY, a large-scale speech dataset collected from actual counseling sessions. The dataset consists of 156 counseling sessions involving 39 therapist-client dyads. The process of speech data collection, subjective ratings (one observer and two client ratings), and transcription are described. An automatic speech and text processing system is developed to locate the time stamps of speaker turns in each session. Examining the relationships among the three subjective ratings suggests that observer and client ratings have no significant correlation, while the client-rated measures are significantly correlated. The intensity similarity between the therapist and the client, measured by the averaged absolute difference of speaker-turn-level intensities, is associated with the psychotherapy outcomes. Recent studies on the acoustic and linguistic characteristics of the CUEMPATHY are introduced.

Forming a reliable judgement of a machine learning (ML) model's appropriateness for an application ecosystem is critical for its responsible use, and requires considering a broad range of factors including harms, benefits, and responsibilities. In practice, however, evaluations of ML models frequently focus on only a narrow range of decontextualized predictive behaviours. We examine the evaluation gaps between the idealized breadth of evaluation concerns and the observed narrow focus of actual evaluations. Through an empirical study of papers from recent high-profile conferences in the Computer Vision and Natural Language Processing communities, we demonstrate a general focus on a handful of evaluation methods. By considering the metrics and test data distributions used in these methods, we draw attention to which properties of models are centered in the field, revealing the properties that are frequently neglected or sidelined during evaluation. By studying these properties, we demonstrate the machine learning discipline's implicit assumption of a range of commitments which have normative impacts; these include commitments to consequentialism, abstractability from context, the quantifiability of impacts, the limited role of model inputs in evaluation, and the equivalence of different failure modes. Shedding light on these assumptions enables us to question their appropriateness for ML system contexts, pointing the way towards more contextualized evaluation methodologies for robustly examining the trustworthiness of ML models

Benchmark contamination has become a significant concern in the LLM evaluation community. Previous Agents-as-an-Evaluator address this issue by involving agents in the generation of questions. Despite their success, the biases in Agents-as-an-Evaluator methods remain largely unexplored. In this paper, we present a theoretical formulation of evaluation bias, providing valuable insights into designing unbiased evaluation protocols. Furthermore, we identify two type of bias in Agents-as-an-Evaluator through carefully designed probing tasks on a minimal Agents-as-an-Evaluator setup. To address these issues, we propose the Unbiased Evaluator, an evaluation protocol that delivers a more comprehensive, unbiased, and interpretable assessment of LLMs.Extensive experiments reveal significant room for improvement in current LLMs. Additionally, we demonstrate that the Unbiased Evaluator not only offers strong evidence of benchmark contamination but also provides interpretable evaluation results.

As language models shift from single-shot answer generation toward multi-step reasoning that retrieves and consumes evidence mid-inference, evaluating the role of individual retrieved items becomes more important. Existing RAG evaluation typically targets final-answer quality, citation faithfulness, or answer-level attribution, but none of these directly targets the intervention-based, per-evidence-item utility view we study here. We introduce CUE-R, a lightweight intervention-based framework for measuring per-evidence-item operational utility in single-shot RAG using shallow observable retrieval-use traces. CUE-R perturbs individual evidence items via REMOVE, REPLACE, and DUPLICATE operators, then measures changes along three utility axes (correctness, proxy-based grounding faithfulness, and confidence error) plus a trace-divergence signal. We also outline an operational evidence-role taxonomy for interpreting intervention outcomes. Experiments on HotpotQA and 2WikiMultihopQA with Qwen-3 8B and GPT-5.2 reveal a consistent pattern: REMOVE and REPLACE substantially harm correctness and grounding while producing large trace shifts, whereas DUPLICATE is often answer-redundant yet not fully behaviorally neutral. A zero-retrieval control confirms that these effects arise from degradation of meaningful retrieval. A two-support ablation further shows that multi-hop evidence items can interact non-additively: removing both supports harms performance far more than either single removal. Our results suggest that answer-only evaluation misses important evidence effects and that intervention-based utility analysis is a practical complement for RAG evaluation.

In this paper, we systematically investigate the mechanisms, scaling behavior, and generalization of physical Recursive self-improvement. We introduce PhysRSI-eval to quantify physical RSI through a five-axis metric: First-Attempt Performance, Efficiency, Success, Budget and Improvement from failure. Different VLMs are evaluated on 18 manipulation tasks under a fixed interaction budget. The analysis of experimental results motivates the Physical Barrel Hypothesis: different capabilities may constrain different parts of improvement. The five-axis metrics also distinguish systems with close public benchmark averages and show that high scores need not come with early or efficient gains. We study observational scaling from two perspectives: predicting the relative performance of VLM agents from their public capabilities, and forecasting later progress from early physical feedback. Together, these analyses ask how quickly and how far an agent can improve. Moreover, we conduct workspace interventions and then analyze what improvement leaves behind. Removing history reduces late-round success, frozen workspaces remain useful on unseen layouts, and three cross-task cases achieve success earlier. The results suggest that scene-specific adaptation and reusable experience can coexist.

Interpretability research on large language models (LLMs) has yielded important insights into model behaviour, yet recurring pitfalls persist: findings that do not generalise, and causal interpretations that outrun the evidence. Our position is that causal inference specifies what constitutes a valid mapping from model activations to invariant high-level structures, the data or assumptions needed to achieve it, and the inferences it can support. Specifically, Pearl's causal hierarchy clarifies what an interpretability study can justify. Observations establish associations between model behaviour and internal components. Interventions (e.g., ablations or activation patching) support claims how these edits affect a behavioural metric (e.g., average change in token probabilities) over a set of prompts. However, counterfactual claims -- i.e., asking what the model output would have been for the same prompt under an unobserved intervention -- remain largely unverifiable without controlled supervision. We show how causal representation learning (CRL) operationalises this hierarchy, specifying which variables are recoverable from activations and under what assumptions. Together, these motivate a diagnostic framework that helps practitioners select methods and evaluations matching claims to evidence such that findings generalise.

Large language models (LLMs) have shown strong performance on isolated psychiatric tasks, including dialogue, diagnosis, and treatment planning, yet existing benchmarks rarely simulate complete psychiatric clinical encounters. We introduce MentalHospital, a virtual evaluation environment for LLM-based psychiatric clinical encounters. MentalHospital instantiates the Subjective Interviewing, Objective Examination, Diagnostic Assessment, and Treatment Planning (S.O.A.P.) workflow, using skill-augmented standardized patients constructed from 1,193 de-identified psychiatric electronic health record (EHR) cases spanning all major ICD-11 categories and 76 disorders. Each encounter is assessed through a dual-track protocol that combines objective comparison against EHR-derived references with subjective assessment of clinical process quality. To scale specialist judgment, we develop MentalEval, five domain-specific evaluators covering communication empathy, interviewing professionalism, clinical-note quality, diagnostic rigor, and treatment appropriateness, trained with rubric-grounded SFT and expert-guided DPO. Survey responses from 22 clinicians support MentalHospital's clinical fidelity (3.88/5), while MentalEval achieves strong expert alignment with an average QWK of 0.944. Benchmarking shows that even the strongest LLM trails clinicians by 37.28 percentage points in objective psychiatric competence, with mental status assessment as a key bottleneck.

AI answer engines increasingly mediate access to domain knowledge by generating responses and citing web sources. We introduce GEO-16, a 16 pillar auditing framework that converts on page quality signals into banded pillar scores and a normalized GEO score G that ranges from 0 to 1. Using 70 product intent prompts, we collected 1,702 citations across three engines (Brave Summary, Google AI Overviews, and Perplexity) and audited 1,100 unique URLs. In our corpus, the engines differed in the GEO quality of the pages they cited, and pillars related to Metadata and Freshness, Semantic HTML, and Structured Data showed the strongest associations with citation. Logistic models with domain clustered standard errors indicate that overall page quality is a strong predictor of citation, and simple operating points (for example, G at least 0.70 combined with at least 12 pillar hits) align with substantially higher citation rates in our data. We report per engine contrasts, vertical effects, threshold analysis, and diagnostics, then translate findings into a practical playbook for publishers. The study is observational and focuses on English language B2B SaaS pages; we discuss limitations, threats to validity, and reproducibility considerations.

Language models can distinguish between testing and deployment phases -- a capability known as evaluation awareness. This has significant safety and policy implications, potentially undermining the reliability of evaluations that are central to AI governance frameworks and voluntary industry commitments. In this paper, we study evaluation awareness in Llama-3.3-70B-Instruct. We show that linear probes can separate real-world evaluation and deployment prompts, suggesting that current models internally represent this distinction. We also find that current safety evaluations are correctly classified by the probes, suggesting that they already appear artificial or inauthentic to models. Our findings underscore the importance of ensuring trustworthy evaluations and understanding deceptive capabilities. More broadly, our work showcases how model internals may be leveraged to support blackbox methods in safety audits, especially for future models more competent at evaluation awareness and deception.

Peer review in machine learning is under growing pressure from rising submission volume and limited reviewer time. Most LLM-based reviewing systems read only the manuscript and generate comments from the paper's own narrative. This makes their outputs sensitive to presentation quality and leaves them weak when the evidence needed for review lies in related work or released code. We present FactReview, an evidence-grounded reviewing system that combines claim extraction, literature positioning, and execution-based claim verification. Given a submission, FactReview identifies major claims and reported results, retrieves nearby work to clarify the paper's technical position, and, when code is available, executes the released repository under bounded budgets to test central empirical claims. It then produces a concise review and an evidence report that assigns each major claim one of five labels: Supported, Supported by the paper, Partially supported, In conflict, or Inconclusive. In a case study on CompGCN, FactReview reproduces results that closely match those reported for link prediction and node classification, yet also shows that the paper's broader performance claim across tasks is not fully sustained: on MUTAG graph classification, the reproduced result is 88.4%, whereas the strongest baseline reported in the paper remains 92.6%. The claim is therefore only partially supported. More broadly, this case suggests that AI is most useful in peer review not as a final decision-maker, but as a tool for gathering evidence and helping reviewers produce more evidence-grounded assessments. The code is public at https://github.com/DEFENSE-SEU/Review-Assistant.

The automation of scientific discovery has been a long-standing goal within the research community, driven by the potential to accelerate knowledge creation. While significant progress has been made using commercial large language models (LLMs) as research assistants or idea generators, the possibility of automating the entire research process with open-source LLMs remains largely unexplored. This paper explores the feasibility of using open-source post-trained LLMs as autonomous agents capable of performing the full cycle of automated research and review, from literature review and manuscript preparation to peer review and paper revision. Our iterative preference training framework consists of CycleResearcher, which conducts research tasks, and CycleReviewer, which simulates the peer review process, providing iterative feedback via reinforcement learning. To train these models, we develop two new datasets, Review-5k and Research-14k, reflecting real-world machine learning research and peer review dynamics. Our results demonstrate that CycleReviewer achieves a 26.89\% improvement in mean absolute error (MAE) over individual human reviewers in predicting paper scores, indicating that LLMs can surpass expert-level performance in research evaluation. In research, the papers generated by the CycleResearcher model achieved a score of 5.36 in simulated peer reviews, surpassing the preprint level of 5.24 from human experts and approaching the accepted paper level of 5.69. This work represents a significant step toward fully automated scientific inquiry, providing ethical safeguards and advancing AI-driven research capabilities. The code, dataset and model weight are released at http://github/minjun-zhu/Researcher.

Natural language is an appealing medium for explaining how large language models process and store information, but evaluating the faithfulness of such explanations is challenging. To help address this, we develop two modes of evaluation for natural language explanations that claim individual neurons represent a concept in a text input. In the observational mode, we evaluate claims that a neuron a activates on all and only input strings that refer to a concept picked out by the proposed explanation E. In the intervention mode, we construe E as a claim that the neuron a is a causal mediator of the concept denoted by E. We apply our framework to the GPT-4-generated explanations of GPT-2 XL neurons of Bills et al. (2023) and show that even the most confident explanations have high error rates and little to no causal efficacy. We close the paper by critically assessing whether natural language is a good choice for explanations and whether neurons are the best level of analysis.

We investigate how looped transformers encode human preference, training lightweight evaluator heads on frozen Ouro-2.6B loop-iteration states on Anthropic HH-RLHF. v2: an erratum is prepended; the original manuscript is unchanged. A post-publication audit found the three headline results inflated by two independent evaluation errors. The 95.2% pairwise evaluator accuracy is a canonical-ordering artifact: the data were correctly split, but the evaluator learned to prefer the first-presented argument; its strict antisymmetrized accuracy on the full 8,552-pair test set is 63.9%. The 84.5% pairwise probe and the below-chance 21.75% pointwise probe were source-item leaks (orientation rows and pair partners crossing the train/test split); corrected pair-disjoint values are 56.5% and 54.2% -- above chance, so the "inverted polarity" finding is withdrawn. The central finding survives at much smaller magnitude: preference is decoded more accurately relationally than pointwise (paired +2.3 points, 95% CI [+1.3, +3.3]), and the antisymmetrized evaluator still beats the linear probe, but no corrected readout rivals end-to-end reward models. The methodological findings stand (constant-output degeneracy, flip test, swap-protocol metric deflation), with one correction: antisymmetrized accuracy, not antisymmetry correlation, certifies relational discrimination. The two errors are mutually invisible -- a split audit cannot see an ordering prior, antisymmetrization cannot see a leak -- so both checks are required. Full audit in the follow-up work (Kirin, 2026, in preparation).

Counterfactual prompting (i.e., perturbing a single factor and measuring output change) is widely used to evaluate things like LLM bias and CoT faithfulness. But in this work we argue that observed effects cannot be attributed to the targeted factor without accounting for baseline ``meaning-preserving'' modifications to text that establish general model sensitivity. This is because every counterfactual edit is a compound treatment that bundles the variable of interest with incidental surface-form variation; this violates treatment variation irrelevance. We observe prediction flip rates on MedQA of 14.9% when we surgically change patient gender. However, this is statistically indistinguishable from the flip rates induced by simply paraphrasing inputs (14.1%). In this case, it would therefore be unwarranted to conclude that the LLM is especially sensitive to patient gender. To account for this and robustly measure the effects of targeted interventions, we propose a framework in which we compare (via statistical testing) differences observed under target interventions to those induced by paraphrasing inputs. We then use this framework to revisit a analysis done on the MedPerturb dataset, which reported evidence of model sensitivity to patient demographics and stylistic cues. We find that these effects largely dissipate when we account for general model sensitivity, with only 5 of 120 tests reaching statistical significance. Applying the same framework to occupational biography classification, we detect clearly significant directional gender bias, showing that the framework identifies real directional effects even when they are small. We evaluate a range of metrics -- aggregate, per-sample distributional, and regression -- and find that per-sample metrics are dramatically more powerful than aggregate metrics and regression powerfully and uniquely characterizes effect direction and magnitude.

Users of music streaming, video streaming, news recommendation, and e-commerce services often engage with content in a sequential manner. Providing and evaluating good sequences of recommendations is therefore a central problem for these services. Prior reweighting-based counterfactual evaluation methods either suffer from high variance or make strong independence assumptions about rewards. We propose a new counterfactual estimator that allows for sequential interactions in the rewards with lower variance in an asymptotically unbiased manner. Our method uses graphical assumptions about the causal relationships of the slate to reweight the rewards in the logging policy in a way that approximates the expected sum of rewards under the target policy. Extensive experiments in simulation and on a live recommender system show that our approach outperforms existing methods in terms of bias and data efficiency for the sequential track recommendations problem.

Commonly used evaluation measures including Recall, Precision, F-Measure and Rand Accuracy are biased and should not be used without clear understanding of the biases, and corresponding identification of chance or base case levels of the statistic. Using these measures a system that performs worse in the objective sense of Informedness, can appear to perform better under any of these commonly used measures. We discuss several concepts and measures that reflect the probability that prediction is informed versus chance. Informedness and introduce Markedness as a dual measure for the probability that prediction is marked versus chance. Finally we demonstrate elegant connections between the concepts of Informedness, Markedness, Correlation and Significance as well as their intuitive relationships with Recall and Precision, and outline the extension from the dichotomous case to the general multi-class case.

Providing constructive feedback to paper authors is a core component of peer review. With reviewers increasingly having less time to perform reviews, automated support systems are required to ensure high reviewing quality, thus making the feedback in reviews useful for authors. To this end, we identify four key aspects of review comments (individual points in weakness sections of reviews) that drive the utility for authors: Actionability, Grounding & Specificity, Verifiability, and Helpfulness. To enable evaluation and development of models assessing review comments, we introduce the RevUtil dataset. We collect 1,430 human-labeled review comments and scale our data with 10k synthetically labeled comments for training purposes. The synthetic data additionally contains rationales, i.e., explanations for the aspect score of a review comment. Employing the RevUtil dataset, we benchmark fine-tuned models for assessing review comments on these aspects and generating rationales. Our experiments demonstrate that these fine-tuned models achieve agreement levels with humans comparable to, and in some cases exceeding, those of powerful closed models like GPT-4o. Our analysis further reveals that machine-generated reviews generally underperform human reviews on our four aspects.

Observability into the decision making of modern AI systems may be required to safely deploy increasingly capable agents. Monitoring the chain-of-thought (CoT) of today's reasoning models has proven effective for detecting misbehavior. However, this "monitorability" may be fragile under different training procedures, data sources, or even continued system scaling. To measure and track monitorability, we propose three evaluation archetypes (intervention, process, and outcome-property) and a new monitorability metric, and introduce a broad evaluation suite. We demonstrate that these evaluations can catch simple model organisms trained to have obfuscated CoTs, and that CoT monitoring is more effective than action-only monitoring in practical settings. We compare the monitorability of various frontier models and find that most models are fairly, but not perfectly, monitorable. We also evaluate how monitorability scales with inference-time compute, reinforcement learning optimization, and pre-training model size. We find that longer CoTs are generally more monitorable and that RL optimization does not materially decrease monitorability even at the current frontier scale. Notably, we find that for a model at a low reasoning effort, we could instead deploy a smaller model at a higher reasoning effort (thereby matching capabilities) and obtain a higher monitorability, albeit at a higher overall inference compute cost. We further investigate agent-monitor scaling trends and find that scaling a weak monitor's test-time compute when monitoring a strong agent increases monitorability. Giving the weak monitor access to CoT not only improves monitorability, but it steepens the monitor's test-time compute to monitorability scaling trend. Finally, we show we can improve monitorability by asking models follow-up questions and giving their follow-up CoT to the monitor.

Treatment effect estimation is of high-importance for both researchers and practitioners across many scientific and industrial domains. The abundance of observational data makes them increasingly used by researchers for the estimation of causal effects. However, these data suffer from biases, from several weaknesses, leading to inaccurate causal effect estimations, if not handled properly. Therefore, several machine learning techniques have been proposed, most of them focusing on leveraging the predictive power of neural network models to attain more precise estimation of causal effects. In this work, we propose a new methodology, named Nearest Neighboring Information for Causal Inference (NNCI), for integrating valuable nearest neighboring information on neural network-based models for estimating treatment effects. The proposed NNCI methodology is applied to some of the most well established neural network-based models for treatment effect estimation with the use of observational data. Numerical experiments and analysis provide empirical and statistical evidence that the integration of NNCI with state-of-the-art neural network models leads to considerably improved treatment effect estimations on a variety of well-known challenging benchmarks.

Claims made by individuals or entities are oftentimes nuanced and cannot be clearly labeled as entirely "true" or "false" -- as is frequently the case with scientific and political claims. However, a claim (e.g., "vaccine A is better than vaccine B") can be dissected into its integral aspects and sub-aspects (e.g., efficacy, safety, distribution), which are individually easier to validate. This enables a more comprehensive, structured response that provides a well-rounded perspective on a given problem while also allowing the reader to prioritize specific angles of interest within the claim (e.g., safety towards children). Thus, we propose ClaimSpect, a retrieval-augmented generation-based framework for automatically constructing a hierarchy of aspects typically considered when addressing a claim and enriching them with corpus-specific perspectives. This structure hierarchically partitions an input corpus to retrieve relevant segments, which assist in discovering new sub-aspects. Moreover, these segments enable the discovery of varying perspectives towards an aspect of the claim (e.g., support, neutral, or oppose) and their respective prevalence (e.g., "how many biomedical papers believe vaccine A is more transportable than B?"). We apply ClaimSpect to a wide variety of real-world scientific and political claims featured in our constructed dataset, showcasing its robustness and accuracy in deconstructing a nuanced claim and representing perspectives within a corpus. Through real-world case studies and human evaluation, we validate its effectiveness over multiple baselines.

Intraoperative anesthesia requires systems to interpret evolving multimodal evidence, recommend timely management, and revise decisions as patient states change, yet existing benchmarks usually isolate perception or single-point reasoning. We introduce AnesTRACE, an evaluation suite comprising AnesTRACE-Bench and AnesTRACE-Eval. Built from public perioperative datasets with anesthesiologist annotation, AnesTRACE-Bench evaluates Intraoperative Perception, Single-point Anesthesia Decision-Making, and Multi-step Anesthesia Decision-Making. AnesTRACE-Eval assesses open-ended responses through anesthesiologist-defined criteria for Clinical Correctness, Evidence Grounding, Task Completeness, and Safety, with Temporal Consistency for multi-step decisions; its domain-specific evaluator is trained by supervised fine-tuning and preference alignment on expert-reviewed judgments. Across more than 30 models, fine-grained visual grounding and intervention selection remain difficult: the leading model reaches only 32.2 mIoU for TEE visual grounding and retains a 17.5\% Major/Critical Safety Error Rate in multi-step management. Evaluator alignment with anesthesiologists improves across both training stages, while the best decision quality is accompanied by a 74.3-second P95 Latency. These results show that aggregate performance alone does not establish safe, timely longitudinal decision-making. We release our code at https://zjudbxai.github.io/AnesTRACE/.

Human-factor analysis typically employs correlation analysis and significance testing to identify relationships between variables. However, these descriptive ('what-is') methods, while effective for identifying associations, are often insufficient for answering causal ('what-if') questions. Their application in such contexts often overlooks confounding and colliding variables, potentially leading to bias and suboptimal or incorrect decisions. We advocate for explicitly distinguishing descriptive from interventional questions in human-factor analysis, and applying causal inference frameworks specifically to these problems to prevent methodological mismatches. This approach disentangles complex variable relationships and enables counterfactual reasoning. Using post-occupancy evaluation (POE) data from the Center for the Built Environment's (CBE) Occupant Survey as a demonstration case, we show how causal discovery reveals intervention hierarchies and directional relationships that traditional associational analysis misses. The systematic distinction between causally associated and independent variables, combined with intervention prioritization capabilities, offers broad applicability to complex human-centric systems, for example, in building science or ergonomics, where understanding intervention effects is critical for optimization and decision-making.

Recovering causal relationships from data is an important problem. Using observational data, one can typically only recover causal graphs up to a Markov equivalence class and additional assumptions or interventional data are needed for complete recovery. In this work, under some standard assumptions, we study causal graph discovery via adaptive interventions with node-dependent interventional costs. For this setting, we show that no algorithm can achieve an approximation guarantee that is asymptotically better than linear in the number of vertices with respect to the verification number; a well-established benchmark for adaptive search algorithms. Motivated by this negative result, we define a new benchmark that captures the worst-case interventional cost for any search algorithm. Furthermore, with respect to this new benchmark, we provide adaptive search algorithms that achieve logarithmic approximations under various settings: atomic, bounded size interventions and generalized cost objectives.

Due to the cumbersome nature of human evaluation and limitations of code-based evaluation, Large Language Models (LLMs) are increasingly being used to assist humans in evaluating LLM outputs. Yet LLM-generated evaluators simply inherit all the problems of the LLMs they evaluate, requiring further human validation. We present a mixed-initiative approach to ``validate the validators'' -- aligning LLM-generated evaluation functions (be it prompts or code) with human requirements. Our interface, EvalGen, provides automated assistance to users in generating evaluation criteria and implementing assertions. While generating candidate implementations (Python functions, LLM grader prompts), EvalGen asks humans to grade a subset of LLM outputs; this feedback is used to select implementations that better align with user grades. A qualitative study finds overall support for EvalGen but underscores the subjectivity and iterative process of alignment. In particular, we identify a phenomenon we dub criteria drift: users need criteria to grade outputs, but grading outputs helps users define criteria. What is more, some criteria appears dependent on the specific LLM outputs observed (rather than independent criteria that can be defined a priori), raising serious questions for approaches that assume the independence of evaluation from observation of model outputs. We present our interface and implementation details, a comparison of our algorithm with a baseline approach, and implications for the design of future LLM evaluation assistants.

We introduce Sci-VBench, a comprehensive benchmark for evaluating knowledge- and reasoning-intensive video generation across scientific domains. It contains 1,253 expert-annotated examples spanning 60 subjects across four core disciplines: Natural Science, Healthcare, Humanities & Social Sciences, and Engineering. Each example requires models to generate temporally rich videos that demand scientific reasoning and knowledge-grounded synthesis, going beyond surface-level visual plausibility. We further establish a rubric-based evaluation protocol. Our analysis shows that, under this protocol, both non-expert human evaluators and MLLM-as-Judge systems can achieve relatively high agreement with expert judgments, supporting reproducible evaluation at scale. We benchmark 16 frontier proprietary and open-source models and find that, while automatic perceptual-quality scores cluster tightly across systems, performance on Prompt Grounding and Scientific and Causal Correctness varies substantially, with a pronounced proprietary-open-source gap. These findings show that advances in visual realism have not yet translated into reliable modeling of scientific and causal dynamics.

Peer review relies on substantive, evidence-based questions, yet current LLMs generate surface-level queries that perform worse than human reviewer questions in expert evaluation. To address this gap, we curate a high-quality dataset of reviewer questions from OpenReview and conduct a human preference study where expert annotators evaluate question-paper pairs across three dimensions: effort, evidence, and grounding. From these annotations, we train IntelliReward, a reward model built from a frozen autoregressive LLM with trainable multi-head transformers. Validated against expert judgments, IntelliReward predicts reviewer-question quality better than API-based SFT baselines and provides scalable evaluation. We apply Decoupled Clip and Dynamic Sampling Policy Optimization (DAPO) with IntelliReward to train IntelliAsk, a question-generation model aligned with human standards of effortful, evidence-based critique. Human evaluations show IntelliAsk generates more grounded, substantive and effortful questions than strong baselines and reduces reliance on first-page content. We also find improvements on reasoning and writing benchmarks, suggesting reviewer-question quality correlates with broader capabilities. Compared to Qwen3-32B, IntelliAsk improves MuSR (68.3 vs 64.7 Acc) and WritingBench (8.31 vs 8.07). We release our code, filtered review dataset, expert annotations, IntelliAsk and IntelliReward to support automatic evaluation of grounding, effort, and evidence in LLM-generated review questions.

Despite the potential of Large Language Models (LLMs) in medicine, they may generate responses lacking supporting evidence or based on hallucinated evidence. While Retrieval Augment Generation (RAG) is popular to address this issue, few studies implemented and evaluated RAG in downstream domain-specific applications. We developed a RAG pipeline with 70,000 ophthalmology-specific documents that retrieve relevant documents to augment LLMs during inference time. In a case study on long-form consumer health questions, we systematically evaluated the responses including over 500 references of LLMs with and without RAG on 100 questions with 10 healthcare professionals. The evaluation focuses on factuality of evidence, selection and ranking of evidence, attribution of evidence, and answer accuracy and completeness. LLMs without RAG provided 252 references in total. Of which, 45.3% hallucinated, 34.1% consisted of minor errors, and 20.6% were correct. In contrast, LLMs with RAG significantly improved accuracy (54.5% being correct) and reduced error rates (18.8% with minor hallucinations and 26.7% with errors). 62.5% of the top 10 documents retrieved by RAG were selected as the top references in the LLM response, with an average ranking of 4.9. The use of RAG also improved evidence attribution (increasing from 1.85 to 2.49 on a 5-point scale, P<0.001), albeit with slight decreases in accuracy (from 3.52 to 3.23, P=0.03) and completeness (from 3.47 to 3.27, P=0.17). The results demonstrate that LLMs frequently exhibited hallucinated and erroneous evidence in the responses, raising concerns for downstream applications in the medical domain. RAG substantially reduced the proportion of such evidence but encountered challenges.

Perception of toxicity evolves over time and often differs between geographies and cultural backgrounds. Similarly, black-box commercially available APIs for detecting toxicity, such as the Perspective API, are not static, but frequently retrained to address any unattended weaknesses and biases. We evaluate the implications of these changes on the reproducibility of findings that compare the relative merits of models and methods that aim to curb toxicity. Our findings suggest that research that relied on inherited automatic toxicity scores to compare models and techniques may have resulted in inaccurate findings. Rescoring all models from HELM, a widely respected living benchmark, for toxicity with the recent version of the API led to a different ranking of widely used foundation models. We suggest caution in applying apples-to-apples comparisons between studies and lay recommendations for a more structured approach to evaluating toxicity over time. Code and data are available at https://github.com/for-ai/black-box-api-challenges.

Peer review is a cornerstone of quality control in scientific publishing. With the increasing workload, the unintended use of `quick' heuristics, referred to as lazy thinking, has emerged as a recurring issue compromising review quality. Automated methods to detect such heuristics can help improve the peer-reviewing process. However, there is limited NLP research on this issue, and no real-world dataset exists to support the development of detection tools. This work introduces LazyReview, a dataset of peer-review sentences annotated with fine-grained lazy thinking categories. Our analysis reveals that Large Language Models (LLMs) struggle to detect these instances in a zero-shot setting. However, instruction-based fine-tuning on our dataset significantly boosts performance by 10-20 performance points, highlighting the importance of high-quality training data. Furthermore, a controlled experiment demonstrates that reviews revised with lazy thinking feedback are more comprehensive and actionable than those written without such feedback. We will release our dataset and the enhanced guidelines that can be used to train junior reviewers in the community. (Code available here: https://github.com/UKPLab/arxiv2025-lazy-review)

AI systems can fail silently. The failure propagates through training loops, evaluation pipelines, and production monitoring stacks until downstream harm makes it visible. This paper introduces evaluation blindness: a measurement function M exhibits evaluation blindness with respect to failure class F when it produces readings indistinguishable from a healthy state while the system is actually failing, with no auxiliary signal flagging the gap. The problem surfaces at two lifecycle stages the literature has treated separately. At training time, reward models are gamed, importance-sampling corrections are silently miscalculated, and benchmark contamination inflates fine-tuning evaluations, all while loss curves look healthy and gradient updates proceed normally. At deployment time, monitoring fails to catch six classes of production failure, including an Operational category that is 100% silent by structural definition. We provide a formal detectability predicate unifying both stages. Four training-time case studies trace concrete breakdowns, including a real implementation bug in TRL PR #6594 where gradients are corrupted as loss decreases normally. A six-class taxonomy validated against 50 real-world incidents from court documents and regulatory filings finds that 53% of verifiable public failures were silent. A failure budget framework ties acceptable failure rates to use-case risk class. The implication is direct: measurement infrastructure is a correctness concern across the full AI lifecycle, not just at evaluation time. Data, code, and taxonomy schema are at https://github.com/priyanka25aug/llm-failure-taxonomy.

Contextual forecasting seeks to improve time series forecasting by incorporating forward-looking information, such as planned events, external conditions, and expert expectations. However, reliable evaluation remains difficult: existing benchmarks often have limited domain coverage, lack standardized history-future splits, or provide text that is not guaranteed to inform the future. We introduce GIFT-Eval-CTX, a benchmark that isolates whether forecasting methods can translate verified future information into accurate numerical trajectories. Built upon GIFT-Eval, GIFT-Eval-CTX contains 10,000 instances across 20 datasets. Each instance provides structured effects derived from the realized future: Macro effects describe horizon-wide behavior, while Micro effects specify localized changes. We construct these annotations with an agentic pipeline combining LLM-based discovery and deterministic code validation. We benchmark TSFMs, frontier LLMs, and tool-using agentic systems. The results show that structured effects reduce aggregate relative MASE by 40.7% for GPT-5.6 Sol and 29.6% for Claude Opus 4.8 relative to history-only prompting. The evaluated MoiraiAgent system achieves the best overall result. We further analyze the contributions of structured effects and agentic systems to performance, and identify opportunities for future research.

Many social and environmental phenomena are associated with macroscopic changes in the built environment, captured by satellite imagery on a global scale and with daily temporal resolution. While widely used for prediction, these images and especially image sequences remain underutilized for causal inference, especially in the context of randomized controlled trials (RCTs), where causal identification is established by design. In this paper, we develop and compare a set of general tools for analyzing Conditional Average Treatment Effects (CATEs) from temporal satellite data that can be applied to any RCT where geographical identifiers are available. Through a simulation study, we analyze different modeling strategies for estimating CATE in sequences of satellite images. We find that image sequence representation models with more parameters generally yield a greater ability to detect heterogeneity. To explore the role of model and data choice in practice, we apply the approaches to two influential RCTs -- Banerjee et al. (2015), a poverty study in Cusco, Peru, and Bolsen et al. (2014), a water conservation experiment in Georgia, USA. We benchmark our image sequence models against image-only, tabular-only, and combined image-tabular data sources, summarizing practical implications for investigators in a multivariate analysis. Land cover classifications over satellite images facilitate interpretation of what image features drive heterogeneity. We also show robustness to data and model choice of satellite-based generalization of the RCT results to larger geographical areas outside the original. Overall, this paper shows how satellite sequence data can be incorporated into the analysis of RCTs, and provides evidence about the implications of data, model, and evaluation metric choice for causal analysis.

Prompt-based probing has been widely used in evaluating the abilities of pretrained language models (PLMs). Unfortunately, recent studies have discovered such an evaluation may be inaccurate, inconsistent and unreliable. Furthermore, the lack of understanding its inner workings, combined with its wide applicability, has the potential to lead to unforeseen risks for evaluating and applying PLMs in real-world applications. To discover, understand and quantify the risks, this paper investigates the prompt-based probing from a causal view, highlights three critical biases which could induce biased results and conclusions, and proposes to conduct debiasing via causal intervention. This paper provides valuable insights for the design of unbiased datasets, better probing frameworks and more reliable evaluations of pretrained language models. Furthermore, our conclusions also echo that we need to rethink the criteria for identifying better pretrained language models. We openly released the source code and data at https://github.com/c-box/causalEval.

If AI models can detect when they are being evaluated, the effectiveness of evaluations might be compromised. For example, models could have systematically different behavior during evaluations, leading to less reliable benchmarks for deployment and governance decisions. We investigate whether frontier language models can accurately classify transcripts based on whether they originate from evaluations or real-world deployment, a capability we call evaluation awareness. To achieve this, we construct a diverse benchmark of 1,000 prompts and transcripts from 61 distinct datasets. These span public benchmarks (e.g., MMLU, SWEBench), real-world deployment interactions, and agent trajectories from scaffolding frameworks (e.g., web-browsing agents). Frontier models clearly demonstrate above-random evaluation awareness (Gemini-2.5-Pro reaches an AUC of 0.83), but do not yet surpass our simple human baseline (AUC of 0.92). Furthermore, both AI models and humans are better at identifying evaluations in agentic settings compared to chat settings. Additionally, we test whether models can identify the purpose of the evaluation. Under multiple-choice and open-ended questioning, AI models far outperform random chance in identifying what an evaluation is testing for. Our results indicate that frontier models already exhibit a substantial, though not yet superhuman, level of evaluation-awareness. We recommend tracking this capability in future models.

Suppose Alice trains an open-weight language model and Bob uses a blackbox derivative of Alice's model to produce text. Can Alice prove that Bob is using her model, either by querying Bob's derivative model (query setting) or from the text alone (observational setting)? We formulate this question as an independence testing problem--in which the null hypothesis is that Bob's model or text is independent of Alice's randomized training run--and investigate it through the lens of palimpsestic memorization in language models: models are more likely to memorize data seen later in training, so we can test whether Bob is using Alice's model using test statistics that capture correlation between Bob's model or text and the ordering of training examples in Alice's training run. If Alice has randomly shuffled her training data, then any significant correlation amounts to exactly quantifiable statistical evidence against the null hypothesis, regardless of the composition of Alice's training data. In the query setting, we directly estimate (via prompting) the likelihood Bob's model gives to Alice's training examples and order; we correlate the likelihoods of over 40 fine-tunes of various Pythia and OLMo base models ranging from 1B to 12B parameters with the base model's training data order, achieving a p-value on the order of at most 1e-8 in all but six cases. In the observational setting, we try two approaches based on estimating 1) the likelihood of Bob's text overlapping with spans of Alice's training examples and 2) the likelihood of Bob's text with respect to different versions of Alice's model we obtain by repeating the last phase (e.g., 1%) of her training run on reshuffled data. The second approach can reliably distinguish Bob's text from as little as a few hundred tokens; the first does not involve any retraining but requires many more tokens (several hundred thousand) to achieve high power.

Longitudinal clinical notes contain rich evidence of how patients evolve over time, but converting this signal into training supervision for clinical prediction remains challenging. We extend Foresight Learning to clinical prediction by converting time-ordered MIMIC-III notes into examples consisting of past patient context, a natural-language question about a possible future event, and a label resolved from later documentation. This process yields 6,900 prediction examples from 702 admissions across medications, procedures, organ support, microbiology, and mortality. A small LoRA adapter trained on these examples improves over the prompted base model, reducing expected calibration error from 0.1269 to 0.0398 and Brier score from 0.199 to 0.145, while slightly outperforming GPT-5 point estimates on held-out questions. The approach enables reusable clinical prediction supervision from longitudinal notes without hand-engineered structured features or endpoint-specific classifiers.

We argue that progress toward AGI is theory limited rather than data or scale limited. Building on the critical rationalism of Popper and Deutsch, we challenge the Platonic Representation Hypothesis. Observationally equivalent worlds can diverge under interventions, so observational adequacy alone cannot guarantee interventional competence. We begin by laying foundations, definitions of knowledge, learning, intelligence, counterfactual competence and AGI, and then analyze the limits of observational learning that motivate an error centric shift. We recast the problem as three questions about how explicit and implicit errors evolve under an agent's actions, which errors are unreachable within a fixed hypothesis space, and how conjecture and criticism expand that space. From these questions we propose Causal Mechanics, a mechanisms first program in which hypothesis space change is a first class operation and probabilistic structure is used when useful rather than presumed. We advance structural principles that make error discovery and correction tractable, including a differential Locality and Autonomy Principle for modular interventions, a gauge invariant form of Independent Causal Mechanisms for separability, and the Compositional Autonomy Principle for analogy preservation, together with actionable diagnostics. The aim is a scaffold for systems that can convert unreachable errors into reachable ones and correct them.

Evaluations of LLM systems routinely average over small prompt sets and report models as a ranked table. We ask how much confidence such a table deserves, using LLM-based prompt-structure inference as the case study: eight open model variants across five families and 8B to 675B parameters, caching disabled, 293 raw intermediate representations persisted. The measured phenomenon is unstable to begin with. Identical calls do not reliably recover identical structure, with mean node-set Jaccard from 0.39 to 0.96 and 72% of prompt-model cells never node-set-perfect. Auditing the evaluation weakens its conclusions further, and this is our main contribution. Under a joint cluster bootstrap over prompts, only the bottom of the ranking is firm: the two least reproducible models hold rank in 99% and 86% of replicates, the middle four in 27% to 48%, and the top two in 68% each, so the table identifies the worst model reliably but does not reliably identify the best. Two equally defensible rules for merging repeated campaigns change four of eight rows and move the study-wide headline by 7 percentage points. Checking the inferred structure against ground-truth annotations shows reproducibility cannot be read as accuracy. And four of the eight endpoints were withdrawn within ten weeks of measurement, so the study as specified can no longer be run. Small-sample LLM evaluations can therefore look far more definitive than their evidence supports. We recommend reporting rank stability, per-cell provenance, executed sensitivity comparisons, raw per-run outputs, and a measurement date alongside any ranking.

Personalized treatment effect estimates are often of interest in high-stakes applications -- thus, before deploying a model estimating such effects in practice, one needs to be sure that the best candidate from the ever-growing machine learning toolbox for this task was chosen. Unfortunately, due to the absence of counterfactual information in practice, it is usually not possible to rely on standard validation metrics for doing so, leading to a well-known model selection dilemma in the treatment effect estimation literature. While some solutions have recently been investigated, systematic understanding of the strengths and weaknesses of different model selection criteria is still lacking. In this paper, instead of attempting to declare a global `winner', we therefore empirically investigate success- and failure modes of different selection criteria. We highlight that there is a complex interplay between selection strategies, candidate estimators and the data used for comparing them, and provide interesting insights into the relative (dis)advantages of different criteria alongside desiderata for the design of further illuminating empirical studies in this context.

Existing benchmarks for Earth science multimodal learning exhibit critical limitations in systematic coverage of geosystem components and cross-sphere interactions, often constrained to isolated subsystems (only in Human-activities sphere or atmosphere) with limited evaluation dimensions (less than 16 tasks). To address these gaps, we introduce OmniEarth-Bench, the first comprehensive multimodal benchmark spanning all six Earth science spheres (atmosphere, lithosphere, Oceansphere, cryosphere, biosphere and Human-activities sphere) and cross-spheres with one hundred expert-curated evaluation dimensions. Leveraging observational data from satellite sensors and in-situ measurements, OmniEarth-Bench integrates 29,779 annotations across four tiers: perception, general reasoning, scientific knowledge reasoning and chain-of-thought (CoT) reasoning. This involves the efforts of 2-5 experts per sphere to establish authoritative evaluation dimensions and curate relevant observational datasets, 40 crowd-sourcing annotators to assist experts for annotations, and finally, OmniEarth-Bench is validated via hybrid expert-crowd workflows to reduce label ambiguity. Experiments on 9 state-of-the-art MLLMs reveal that even the most advanced models struggle with our benchmarks, where none of them reach 35\% accuracy. Especially, in some cross-spheres tasks, the performance of leading models like GPT-4o drops to 0.0\%. OmniEarth-Bench sets a new standard for geosystem-aware AI, advancing both scientific discovery and practical applications in environmental monitoring and disaster prediction. The dataset, source code, and trained models were released.

We study the problem of model selection in causal inference, specifically for the case of conditional average treatment effect (CATE) estimation under binary treatments. Unlike model selection in machine learning, there is no perfect analogue of cross-validation as we do not observe the counterfactual potential outcome for any data point. Towards this, there have been a variety of proxy metrics proposed in the literature, that depend on auxiliary nuisance models estimated from the observed data (propensity score model, outcome regression model). However, the effectiveness of these metrics has only been studied on synthetic datasets as we can access the counterfactual data for them. We conduct an extensive empirical analysis to judge the performance of these metrics introduced in the literature, and novel ones introduced in this work, where we utilize the latest advances in generative modeling to incorporate multiple realistic datasets. Our analysis suggests novel model selection strategies based on careful hyperparameter tuning of CATE estimators and causal ensembling.

We present initial results of a forthcoming benchmark for evaluating LLM agents on white-collar tasks of economic value. We evaluate agents on real-world "messy" open-web research tasks of the type that are routine in finance and consulting. In doing so, we lay the groundwork for an LLM agent evaluation suite where good performance directly corresponds to a large economic and societal impact. We built and tested several agent architectures with o1-preview, GPT-4o, Claude-3.5 Sonnet, Llama 3.1 (405b), and GPT-4o-mini. On average, LLM agents powered by Claude-3.5 Sonnet and o1-preview substantially outperformed agents using GPT-4o, with agents based on Llama 3.1 (405b) and GPT-4o-mini lagging noticeably behind. Across LLMs, a ReAct architecture with the ability to delegate subtasks to subagents performed best. In addition to quantitative evaluations, we qualitatively assessed the performance of the LLM agents by inspecting their traces and reflecting on their observations. Our evaluation represents the first in-depth assessment of agents' abilities to conduct challenging, economically valuable analyst-style research on the real open web.

The way people respond to messaging from public health organizations on social media can provide insight into public perceptions on critical health issues, especially during a global crisis such as COVID-19. It could be valuable for high-impact organizations such as the US Centers for Disease Control and Prevention (CDC) or the World Health Organization (WHO) to understand how these perceptions impact reception of messaging on health policy recommendations. We collect two datasets of public health messages and their responses from Twitter relating to COVID-19 and Vaccines, and introduce a predictive method which can be used to explore the potential reception of such messages. Specifically, we harness a generative model (GPT-2) to directly predict probable future responses and demonstrate how it can be used to optimize expected reception of important health guidance. Finally, we introduce a novel evaluation scheme with extensive statistical testing which allows us to conclude that our models capture the semantics and sentiment found in actual public health responses.

Peer review underpins scientific progress, but it is increasingly strained by reviewer shortages and growing workloads. Large Language Models (LLMs) can automatically draft reviews now, but determining whether LLM-generated reviews are trustworthy requires systematic evaluation. Researchers have evaluated LLM reviews at either surface-level (e.g., BLEU and ROUGE) or content-level (e.g., specificity and factual accuracy). Yet it remains uncertain whether LLM-generated reviews attend to the same critical facets that human experts weigh -- the strengths and weaknesses that ultimately drive an accept-or-reject decision. We introduce a focus-level evaluation framework that operationalizes the focus as a normalized distribution of attention across predefined facets in paper reviews. Based on the framework, we developed an automatic focus-level evaluation pipeline based on two sets of facets: target (e.g., problem, method, and experiment) and aspect (e.g., validity, clarity, and novelty), leveraging 676 paper reviews (https://figshare.com/s/d5adf26c802527dd0f62) from OpenReview that consists of 3,657 strengths and weaknesses identified from human experts. The comparison of focus distributions between LLMs and human experts showed that the off-the-shelf LLMs consistently have a more biased focus towards examining technical validity while significantly overlooking novelty assessment when criticizing papers.

Commonsense causality reasoning (CCR) aims at identifying plausible causes and effects in natural language descriptions that are deemed reasonable by an average person. Although being of great academic and practical interest, this problem is still shadowed by the lack of a well-posed theoretical framework; existing work usually relies on deep language models wholeheartedly, and is potentially susceptible to confounding co-occurrences. Motivated by classical causal principles, we articulate the central question of CCR and draw parallels between human subjects in observational studies and natural languages to adopt CCR to the potential-outcomes framework, which is the first such attempt for commonsense tasks. We propose a novel framework, ROCK, to Reason O(A)bout Commonsense K(C)ausality, which utilizes temporal signals as incidental supervision, and balances confounding effects using temporal propensities that are analogous to propensity scores. The ROCK implementation is modular and zero-shot, and demonstrates good CCR capabilities.

To effectively optimize and personalize treatments, it is necessary to investigate the heterogeneity of treatment effects. With the wide range of users being treated over many online controlled experiments, the typical approach of manually investigating each dimension of heterogeneity becomes overly cumbersome and prone to subjective human biases. We need an efficient way to search through thousands of experiments with hundreds of target covariates and hundreds of breakdown dimensions. In this paper, we propose a systematic paradigm for detecting, surfacing and characterizing heterogeneous treatment effects. First, we detect if treatment effect variation is present in an experiment, prior to specifying any breakdowns. Second, we surface the most relevant dimensions for heterogeneity. Finally, we characterize the heterogeneity beyond just the conditional average treatment effects (CATE) by studying the conditional distributions of the estimated individual treatment effects. We show the effectiveness of our methods using simulated data and empirical studies.

Using large language models (LLMs) to evaluate text quality has recently gained popularity. Some prior works explore the idea of using LLMs for evaluation, while they differ in some details of the evaluation process. In this paper, we analyze LLM evaluation (Chiang and Lee, 2023) and G-Eval (Liu et al., 2023), and we discuss how those details in the evaluation process change how well the ratings given by LLMs correlate with human ratings. We find that the auto Chain-of-Thought (CoT) used in G-Eval does not always make G-Eval more aligned with human ratings. We also show that forcing the LLM to output only a numeric rating, as in G-Eval, is suboptimal. Last, we reveal that asking the LLM to explain its own ratings consistently improves the correlation between the ChatGPT and human ratings and pushes state-of-the-art (SoTA) correlations on two meta-evaluation datasets.

How many experimental studies would have come to different conclusions had they been run on larger samples? I show how to estimate the expected number of statistically significant results that a set of experiments would have reported had their sample sizes all been counterfactually increased. The proposed deconvolution estimator is asymptotically normal and adjusts for publication bias. Unlike related methods, this approach requires no assumptions of any kind about the distribution of true intervention treatment effects and allows for point masses. Simulations find good coverage even when the t-score is only approximately normal. An application to randomized trials (RCTs) published in economics journals finds that doubling every sample would increase the power of t-tests by 7.2 percentage points on average. This effect is smaller than for non-RCTs and comparable to systematic replications in laboratory psychology where previous studies enabled more accurate power calculations. This suggests that RCTs are on average relatively insensitive to sample size increases. Research funders who wish to raise power should generally consider sponsoring better-measured and higher quality experiments -- rather than only larger ones.

Political scientists are rapidly adopting large language models (LLMs) for text annotation, yet the sensitivity of annotation results to implementation choices remains poorly understood. Most evaluations test a single model or configuration; how model choice, model size, learning approach, and prompt style interact, and whether popular "best practices" survive controlled comparison, are largely unexplored. We present a controlled evaluation of these pipeline choices, testing six open-weight models across four political science annotation tasks under identical quantisation, hardware, and prompt-template conditions. Our central finding is methodological: interaction effects dominate main effects, so seemingly reasonable pipeline choices can become consequential researcher degrees of freedom. No single model, prompt style, or learning approach is uniformly superior, and the best-performing model varies across tasks. Two corollaries follow. First, model size is an unreliable guide both to cost and to performance: cross-family efficiency differences are so large that some larger models are less resource-intensive than much smaller alternatives, while within model families mid-range variants often match or exceed larger counterparts. Second, widely recommended prompt engineering techniques yield inconsistent and sometimes negative effects on annotation performance. We use these benchmark results to develop a validation-first framework - with a principled ordering of pipeline decisions, guidance on prompt freezing and held-out evaluation, reporting standards, and open-source tools - to help researchers navigate this decision space transparently.

In observational settings where treatment and outcome share unmeasured confounders but an observed mediator remains unconfounded, the front-door (FD) adjustment identifies causal effects through the mediator. We study the heterogeneous treatment effect (HTE) under FD identification and introduce two debiased learners: FD-DR-Learner and FD-R-Learner. Both attain fast, quasi-oracle rates (i.e., performance comparable to an oracle that knows the nuisances) even when nuisance functions converge as slowly as n^-1/4. We provide error analyses establishing debiasedness and demonstrate robust empirical performance in synthetic studies and a real-world case study of primary seat-belt laws using Fatality Analysis Reporting System (FARS) dataset. Together, these results indicate that the proposed learners deliver reliable and sample-efficient HTE estimates in FD scenarios. The implementation is available at https://github.com/yonghanjung/FD-CATE. Keywords: Front-door adjustment; Heterogeneous treatment effects; Debiased learning; Quasi-oracle rates; Causal inference.

As Large Reasoning Models (LRMs) are increasingly deployed, auditing their chain-of-thought (CoT) traces for safety becomes critical. Recent work has reported that monitorability--the degree to which CoT faithfully and informatively reflects internal computation--can appear as a "free gift" during the early stages of Reinforcement Learning with Verifiable Rewards (RLVR). We make this observation concrete through a systematic evaluation across model families and training domains. Our results show that this effect is not universal: monitorability improvements are strongly data-dependent. In particular, we demonstrate the critical role of data diversity and instruction-following data during RLVR training. We further show that monitorability is orthogonal to capability--improvements in reasoning performance do not imply increased transparency. Through mechanistic analysis, we attribute monitorability gains primarily to response distribution sharpening (entropy reduction) and increased attention to the prompt, rather than stronger causal reliance on reasoning traces. We also reveal how monitorability dynamics vary with controlled training and evaluation difficulty. Together, these findings provide a holistic view of how monitorability emerges under RLVR, clarifying when gains are likely to occur and when they are not.

Health economic evaluations are sensitive to the choice of analytical perspective (e.g., health system vs. societal). While guidelines often recommend specific perspectives, the uncertainty associated with this choice - and the potential decision discordance it creates - is rarely quantified. We present vop_poc_nz, a Python package that implements a framework for Distributional Cost-Effectiveness Analysis (DCEA) and operationalizes the quantification of perspective uncertainty through the Value of Perspective (VoP) metric. The package provides tools for Markov modeling, probabilistic sensitivity analysis, value of information analysis, and equity impact assessment. Unlike existing tools that treat perspective as a fixed input, vop_poc_nz allows for the simultaneous evaluation of multiple perspectives. This enables decision-makers to estimate the opportunity cost of perspective misalignment. We demonstrate the package's capabilities using case studies from Aotearoa New Zealand.

Evaluation efforts such as TREC, CLEF, NTCIR and FIRE, alongside public leaderboard such as MS MARCO, are intended to encourage research and track our progress, addressing big questions in our field. However, the goal is not simply to identify which run is "best", achieving the top score. The goal is to move the field forward by developing new robust techniques, that work in many different settings, and are adopted in research and practice. This paper uses the MS MARCO and TREC Deep Learning Track as our case study, comparing it to the case of TREC ad hoc ranking in the 1990s. We show how the design of the evaluation effort can encourage or discourage certain outcomes, and raising questions about internal and external validity of results. We provide some analysis of certain pitfalls, and a statement of best practices for avoiding such pitfalls. We summarize the progress of the effort so far, and describe our desired end state of "robust usefulness", along with steps that might be required to get us there.

We consider the task of evaluating policies of algorithmic resource allocation through randomized controlled trials (RCTs). Such policies are tasked with optimizing the utilization of limited intervention resources, with the goal of maximizing the benefits derived. Evaluation of such allocation policies through RCTs proves difficult, notwithstanding the scale of the trial, because the individuals' outcomes are inextricably interlinked through resource constraints controlling the policy decisions. Our key contribution is to present a new estimator leveraging our proposed novel concept, that involves retrospective reshuffling of participants across experimental arms at the end of an RCT. We identify conditions under which such reassignments are permissible and can be leveraged to construct counterfactual trials, whose outcomes can be accurately ascertained, for free. We prove theoretically that such an estimator is more accurate than common estimators based on sample means -- we show that it returns an unbiased estimate and simultaneously reduces variance. We demonstrate the value of our approach through empirical experiments on synthetic, semi-synthetic as well as real case study data and show improved estimation accuracy across the board.

Off-policy evaluation (OPE) is a method for estimating the return of a target policy using some pre-collected observational data generated by a potentially different behavior policy. In some cases, there may be unmeasured variables that can confound the action-reward or action-next-state relationships, rendering many existing OPE approaches ineffective. This paper develops an instrumental variable (IV)-based method for consistent OPE in confounded Markov decision processes (MDPs). Similar to single-stage decision making, we show that IV enables us to correctly identify the target policy's value in infinite horizon settings as well. Furthermore, we propose an efficient and robust value estimator and illustrate its effectiveness through extensive simulations and analysis of real data from a world-leading short-video platform.

Contextual bandit algorithms have become popular for online recommendation systems such as Digg, Yahoo! Buzz, and news recommendation in general. Offline evaluation of the effectiveness of new algorithms in these applications is critical for protecting online user experiences but very challenging due to their "partial-label" nature. Common practice is to create a simulator which simulates the online environment for the problem at hand and then run an algorithm against this simulator. However, creating simulator itself is often difficult and modeling bias is usually unavoidably introduced. In this paper, we introduce a replay methodology for contextual bandit algorithm evaluation. Different from simulator-based approaches, our method is completely data-driven and very easy to adapt to different applications. More importantly, our method can provide provably unbiased evaluations. Our empirical results on a large-scale news article recommendation dataset collected from Yahoo! Front Page conform well with our theoretical results. Furthermore, comparisons between our offline replay and online bucket evaluation of several contextual bandit algorithms show accuracy and effectiveness of our offline evaluation method.

Large language models (LLMs) have recently been applied to forecasting tasks, with some works claiming these systems match or exceed human performance. In this paper, we argue that, as a community, we should be careful about such conclusions as evaluating LLM forecasters presents unique challenges. We identify two broad categories of issues: (1) difficulty in trusting evaluation results due to many forms of temporal leakage, and (2) difficulty in extrapolating from evaluation performance to real-world forecasting. Through systematic analysis and concrete examples from prior work, we demonstrate how evaluation flaws can raise concerns about current and future performance claims. We argue that more rigorous evaluation methodologies are needed to confidently assess the forecasting abilities of LLMs.

Low latent prediction error does not establish that a world model distinguishes the consequences of its actions. We introduce an evaluation protocol that traces the same intervention through simulator state, raster observations, target embeddings, and predictor outputs. Exact simulator-state forks in a controlled deformable-physics testbed reveal distinct bottlenecks. Changed commands alter particle motion, yet 41.5% of one-step raster pairs are identical. Observation loss is not the whole explanation: among 579 high-visibility counterfactuals, median predictor-to-target response is 0.0051 and 0.0217 across two seeds, falling to 0.0027 and 0.0116 after variance normalization. An isotropic state perturbation matched to the target counterfactual embedding shift produces 190x and 53x larger predictor changes on the same visible pairs, isolating action-path under-use rather than a dead or globally shrunk predictor. MSE-only training gives 8.36x lower 10-step latent error in matched seeds, but in spectrally concentrated spaces; one VICReg target encoder is also strongly concentrated, so neither error nor rank alone certifies physical state. Finally, stiffness remains near chance even from full-resolution rasters and mechanical state while privileged material parameters decode perfectly, indicating weak identifiability under this excitation rather than encoder discard. These results motivate auditing physical effect, observation visibility, representation geometry, and action dependence separately.

The large language model (LLM)-as-judge paradigm has been used to meet the demand for a cheap, reliable, and fast evaluation of model outputs during AI system development and post-deployment monitoring. While judge models -- LLMs finetuned to specialize in assessing and critiquing model outputs -- have been touted as general purpose evaluators, they are typically evaluated only on non-contextual scenarios, such as instruction following. The omission of contextual settings -- those where external information is used as context to generate an output -- is surprising given the increasing prevalence of retrieval-augmented generation (RAG) and summarization use cases. Contextual assessment is uniquely challenging, as evaluation often depends on practitioner priorities, leading to conditional evaluation criteria (e.g., comparing responses based on factuality and then considering completeness if they are equally factual). To address the gap, we propose ContextualJudgeBench, a judge benchmark with 2,000 challenging response pairs across eight splits inspired by real-world contextual evaluation scenarios. We build our benchmark with a multi-pronged data construction pipeline that leverages both existing human annotations and model-based perturbations. Our comprehensive study across 11 judge models and 9 general purpose models, reveals that the contextual information and its assessment criteria present a significant challenge to even state-of-the-art models. For example, OpenAI's o1, the best-performing model, barely reaches 55% consistent accuracy.

For observational studies, we study the sensitivity of causal inference when treatment assignments may depend on unobserved confounders. We develop a loss minimization approach for estimating bounds on the conditional average treatment effect (CATE) when unobserved confounders have a bounded effect on the odds ratio of treatment selection. Our approach is scalable and allows flexible use of model classes in estimation, including nonparametric and black-box machine learning methods. Based on these bounds for the CATE, we propose a sensitivity analysis for the average treatment effect (ATE). Our semi-parametric estimator extends/bounds the augmented inverse propensity weighted (AIPW) estimator for the ATE under bounded unobserved confounding. By constructing a Neyman orthogonal score, our estimator of the bound for the ATE is a regular root-n estimator so long as the nuisance parameters are estimated at the o_p(n^{-1/4}) rate. We complement our methodology with optimality results showing that our proposed bounds are tight in certain cases. We demonstrate our method on simulated and real data examples, and show accurate coverage of our confidence intervals in practical finite sample regimes with rich covariate information.

Large language models (LLMs) are now used for academic expert recommendation. Existing audits typically evaluate such recommendations in isolation, ignoring end-user inference-time interventions. Thus, it remains unclear whether failures (e.g., refusals, hallucinations, uneven coverage) stem from model choice or deployment decisions. We introduce LLMScholarBench, a benchmark for auditing LLM-based scholar recommendation that jointly evaluates model infrastructure and end-user interventions across multiple tasks. LLMScholarBench measures technical quality and social representation using nine metrics. We instantiate the benchmark in physics expert recommendation and audit 22 LLMs under temperature variation, representation-constrained prompting, and retrieval-augmented generation (RAG) via web search. Our results show that each intervention entails distinct tradeoffs. Higher temperature degrades validity, consistency, and factuality. Representation-constrained prompting improves diversity at the expense of factuality, while RAG primarily improves technical quality while reducing diversity and parity. Overall, end-user interventions reshape trade-offs rather than providing uniform gains. LLMScholarBench makes all these dynamics auditable across models and interventions in LLM-based scholar recommendations.

Stronger AI agents do not automatically produce better organizations: teams must also learn which work arrangements to retain and when to reconsider them. We propose a modeling specification for recursive organization improvement and evaluate it through an executable checker, a public-record mapping, and controlled simulation. The specification connects actor-visible histories, organizational memory, decision rights, and evidence-carrying change contracts. The mechanism study crosses six decision rules, three memory conditions, and three task environments under fixed resource ceilings. In a stationary environment, cumulative evidence raises balanced evaluation's normalized net value per task from 0.45224 to 0.48007. Repeated reassessment's disadvantage relative to this comparator falls from 0.01702 with reset evidence to 0.00007 with cumulative evidence. A reversal of the best workflow reveals the opposite cost: indefinite retention delays adaptation, while a finite window restores eventual performance at a transition cost. In exploratory controls, matching trial acquisition and label reuse reduces the apparent reassessment gain from 0.00607 to 0.00191. Program replacement adds no stable benefit across the tested reversal times. The study identifies evidence acquisition, reuse, and timely updating as mechanisms that must be separated from evaluator replacement when assessing organizational improvement.

The teaching career shares common global characteristics, such as internal promotion, performance evaluation, recruitment of top candidates, continuous training, specialization, and peer learning. This study aims to describe the factors associated with the value placed on the teaching career in Peru. A total of 28217 public school teachers were analyzed using data from the 2020 National Teacher Survey. A variable measuring the "value of the teaching career" was constructed using eight indicators and categorized as low, medium, or high. Another variable, vision of the future, was classified as pessimistic, conformist, or optimistic. This observational, cross-sectional, and analytical study included variables related to in-service training, working conditions, professional recognition, and sociodemographic characteristics. Among the teachers surveyed, 45.8 % expressed an optimistic outlook on the future of the profession, 48 % held a conformist view, and only 6.2 % reported a pessimistic perspective. A generalized linear model revealed that the value placed on the teaching career was significantly associated with male gender (p = 0.002), a professional career (p < 0.001), an optimistic outlook (p = 0.033), and working at the primary level (p < 0.001). It was concluded that Peruvian teachers predominantly hold conformist or optimistic views of their profession. This highlights the need to reinforce merit-based advancement, competency-based training, intrinsic motivation, and ongoing professional development

We present OpenReviewer, an open-source system for generating high-quality peer reviews of machine learning and AI conference papers. At its core is Llama-OpenReviewer-8B, an 8B parameter language model specifically fine-tuned on 79,000 expert reviews from top conferences. Given a PDF paper submission and review template as input, OpenReviewer extracts the full text, including technical content like equations and tables, and generates a structured review following conference-specific guidelines. Our evaluation on 400 test papers shows that OpenReviewer produces considerably more critical and realistic reviews compared to general-purpose LLMs like GPT-4 and Claude-3.5. While other LLMs tend toward overly positive assessments, OpenReviewer's recommendations closely match the distribution of human reviewer ratings. The system provides authors with rapid, constructive feedback to improve their manuscripts before submission, though it is not intended to replace human peer review. OpenReviewer is available as an online demo and open-source tool.

AI evaluations are widely used for testing and understanding progress. However, the diverse evaluators bring with them inconsistencies that challenge analysis and comparison. First, results are saved in incompatible formats, scattered across leaderboards, papers, blog posts, evaluation harness logs, and custom repositories. Second, results are created by different evaluation frameworks, which produce divergent scores for nominally identical evaluations and record metadata inconsistently, hindering comparison, cross-community evaluation science, cost reduction, and reuse. We introduce Every Eval Ever, the first shared schema and community-crowdsourced repository for AI evaluation results. The schema standardizes how evaluations are represented in a unified, single JSON document. It is source-agnostic by design, ingesting results from evaluation harnesses and papers alike, and optionally stores per-instance outputs for fine-grained analysis. We contribute: (i) a community-governed metadata schema with a companion instance-level schema, the first standardization effort of its kind; (ii) automatic converters from popular formats, evaluation harnesses, and leaderboards to the unified schema; and (iii) a crowdsourced community database hosted on Hugging Face, currently spanning to date 22,235 models, 2,273 unique benchmarks, and 31 evaluation formats.

Automatic evaluation with large language models, commonly known as LLM-as-a-judge, is now standard across reasoning and alignment tasks. Despite evaluating many samples in deployment, these evaluators typically (i) treat each case independently, missing the opportunity to accumulate experience, and (ii) rely on a single fixed prompt for all cases, neglecting the need for sample-specific evaluation criteria. We introduce Learning While Evaluating (LWE), a framework that allows evaluators to improve sequentially at inference time without requiring training or validation sets. LWE maintains an evolving meta-prompt that (i) produces sample-specific evaluation instructions and (ii) refines itself through self-generated feedback. Furthermore, we propose Selective LWE, which updates the meta-prompt only on self-inconsistent cases, focusing computation where it matters most. This selective approach retains the benefits of sequential learning while being far more cost-effective. Across two pairwise comparison benchmarks, Selective LWE outperforms strong baselines, empirically demonstrating that evaluators can improve during sequential testing with a simple selective update, learning most from the cases they struggle with.

Quantifying and assessing urban greenery is consequential for planning and development, reflecting the everlasting importance of green spaces for multiple climate and well-being dimensions of cities. Evaluation can be broadly grouped into objective (e.g., measuring the amount of greenery) and subjective (e.g., polling the perception of people) approaches, which may differ -- what people see and feel about how green a place is might not match the measurements of the actual amount of vegetation. In this work, we advance the state of the art by measuring such differences and explaining them through human, geographic, and spatial dimensions. The experiments rely on contextual information extracted from street view imagery and a comprehensive urban visual perception survey collected from 1,000 people across five countries with their extensive demographic and personality information. We analyze the discrepancies between objective measures (e.g., Green View Index (GVI)) and subjective scores (e.g., pairwise ratings), examining whether they can be explained by a variety of human and visual factors such as age group and spatial variation of greenery in the scene. The findings reveal that such discrepancies are comparable around the world and that demographics and personality do not play a significant role in perception. Further, while perceived and measured greenery correlate consistently across geographies (both where people and where imagery are from), where people live plays a significant role in explaining perceptual differences, with these two, as the top among seven, features that influences perceived greenery the most. This location influence suggests that cultural, environmental, and experiential factors substantially shape how individuals observe greenery in cities.

Existing benchmarks for scientific data analysis evaluate LLMs primarily on code execution or workflow completion, overlooking that scientific analysis serves to support distinct types of scientific claims: hypothesis exploration, statistical inference, mechanistic explanation, each with different assumptions and validity criteria. We introduce SDABench, a benchmark that reorganizes evaluation around six capabilities (descriptive, exploratory, inferential, predictive, causal, and mechanistic) across five domains (Biology, Chemistry, Environment, Geography, Physics). SDABench comprises 527 real-data instances (SDA-Real) and 6000 synthetic instances (SDA-Synth), each in both multiple-choice and open-ended formats, constructed through an automated pipeline. Evaluating 15 representative LLMs, we find that models handle descriptive analysis well but degrade sharply on tasks requiring assumption selection, latent-process modeling, or mechanistic reasoning. SDABench further provides a five-stage error analysis framework that locates where LLMs fail: more advanced models more reliably identify the relevant scope and variables, but still struggle to select appropriate analytical procedures, model variable relationships, and draw valid conclusions.

Large Language Models (LLMs) are increasingly used for clinical decision support, where hallucinations and unsafe suggestions may pose direct risks to patient safety. These risks are particularly challenging as they often manifest as subtle clinical errors that evade detection by generic metrics, while expert-authored fine-grained rubrics remain costly to construct and difficult to scale. In this paper, we propose a retrieval-augmented multi-agent framework designed to automate the generation of instance-specific evaluation rubrics. Our approach grounds evaluation in authoritative medical evidence by decomposing retrieved content into atomic facts and synthesizing them with user interaction constraints to form verifiable, fine-grained evaluation criteria. Evaluated on HealthBench, our framework achieves a Clinical Intent Alignment (CIA) score of 60.12%, a statistically significant improvement over the GPT-4o baseline (55.16%). In discriminative tests, our rubrics yield a mean score delta (μ_Δ = 8.658) and an AUROC of 0.977, nearly doubling the quality separation achieved by GPT-4o baseline (4.972). Beyond evaluation, our rubrics effectively guide response refinement, improving quality by 9.2% (from 59.0% to 68.2%). This provides a scalable and transparent foundation for both evaluating and improving medical LLMs. The code is available at https://anonymous.4open.science/r/Automated-Rubric-Generation-AF3C/.

Generating professional scholarly content, such as peer reviews and rebuttals, requires an intricate synergy between domain reasoning and factual grounding. This work presents a comprehensive framework for the development and evaluation of specialized scholarly agents, InternReviewer and InternAdvocate. We first establish a large-scale, high-quality scholarly dataset and integrate a high-efficiency arXiv retrieval tool to enable active evidence gathering. To optimize these agents, we implement an agentic Reinforcement Learning (RL) paradigm driven by a unified objective metric and reward system. This system avoids the biases of subjective model-based judging by employing multi-dimensional criteria, including reference-anchored semantic alignment, structural compliance, and a strict verification mechanism that cross-checks citations against real-time interaction logs to eliminate hallucinations. Experimental results demonstrate that agents trained within this closed-loop framework exhibit significant improvements in reasoning depth and citation accuracy.

LLM-as-judge evaluation has become the de facto standard for scaling model assessment, but the practice is statistically unsound: uncalibrated scores can invert preferences, naive confidence intervals on uncalibrated scores achieve near-0% coverage, and importance-weighted estimators collapse under limited overlap despite high effective sample size (ESS). We introduce Causal Judge Evaluation (CJE), a framework that fixes all three failures. On n=4,961 Chatbot Arena prompts (after filtering from 5k), CJE achieves 99% pairwise ranking accuracy at full sample size (94% averaged across configurations), matching oracle quality, at 14x lower cost (for ranking 5 policies) by calibrating a 16x cheaper judge on just 5% oracle labels (~250 labels). CJE combines three components: (i) AutoCal-R, reward calibration via mean-preserving isotonic regression; (ii) SIMCal-W, weight stabilization via stacking of S-monotone candidates; and (iii) Oracle-Uncertainty Aware (OUA) inference that propagates calibration uncertainty into confidence intervals. We formalize the Coverage-Limited Efficiency (CLE) diagnostic, which explains why IPS-style estimators fail even when ESS exceeds 90%: the logger rarely visits regions where target policies concentrate. Key findings: SNIPS inverts rankings even with reward calibration (38% pairwise, negative Kendall's tau) due to weight instability; calibrated IPS remains near-random (47%) despite weight stabilization, consistent with CLE; OUA improves coverage from near-0% to ~86% (Direct) and ~96% (stacked-DR), where naive intervals severely under-cover.