Daily Papers

Cognitive diagnosis (CD) is a fundamental task in intelligent education that profiles learner proficiency over knowledge concepts. In real-world learning platforms, newly added items continually introduce previously unseen concepts, necessitating dynamic expansion of the underlying concept space. Yet existing incremental CD models assume a fixed concept space, allowing gradients from new items to overwrite historical pathways and induce catastrophic forgetting. More critically, these methods rely solely on soft constraints to preserve historical diagnoses. Such constraints may fail to satisfy the requirement of diagnostic invariance after incremental updates, a requirement known as psychometric consistency in cognitive diagnosis. Therefore, we propose CLEAN (Continual Learning with Expandable and Architecturally Isolated Networks), a novel incremental CD framework supporting concept-space expansion while providing structural guarantees for pointwise invariance of historical diagnoses. Specifically, CLEAN first introduces a strict topological bipartition protocol, freezes historical diagnostic functions and applies deterministic orthogonal column masking to sever gradient interference. Second, to accommodate concept expansion, expandable full-rank branches with micro-variance initialization are deployed to learn novel concepts. Finally, to verify that this architectural design achieves invariance by construction, we formalize Representation Drift (RD) to quantify the perturbation of historical traits. Extensive experiments on three large-scale educational datasets demonstrate that CLEAN achieves zero RD, preserving old-item metrics identically to static anchors through architectural isolation while remaining competitive with or superior to strong continual-learning baselines on new items.

We examine whether human psychometric questionnaires can serve as reliable tools for characterizing and predicting LLM behavior in everyday user interactions. We analyze eight open-source LLMs by comparing their value and personality profiles derived from two different methods: Likert self-reports on established questionnaires (PVQ-40/21 and BFI-44/10) and generation probabilities over value-laden responses to everyday user queries. The two profiles diverge substantially. Within-construct item consistency, often cited as evidence of stable LLM dispositions, disappears in generation probabilities. We attribute this gap to the fact that explicit lexical cues in established questionnaire items allow models to recognize the target construct and respond in alignment-consistent, socially desirable ways, whereas realistic user queries provide no such cues. In addition, demographic persona prompts shift models' responses to human questionnaires in ways consistent with real human patterns, but no such shifts appear in the generation probabilities of responses to realistic user queries, showing their limited ability to simulate the behaviors of target demographics in real-world user interactions. Overall, our study shows that human psychometric questionnaires are insufficient tools for predicting LLM behavior and suggests generation-based profiling as a more accurate measure.

The versatility of Large Language Models (LLMs) on natural language understanding tasks has made them popular for research in social sciences. To properly understand the properties and innate personas of LLMs, researchers have performed studies that involve using prompts in the form of questions that ask LLMs about particular opinions. In this study, we take a cautionary step back and examine whether the current format of prompting LLMs elicits responses in a consistent and robust manner. We first construct a dataset that contains 693 questions encompassing 39 different instruments of persona measurement on 115 persona axes. Additionally, we design a set of prompts containing minor variations and examine LLMs' capabilities to generate answers, as well as prompt variations to examine their consistency with respect to content-level variations such as switching the order of response options or negating the statement. Our experiments on 17 different LLMs reveal that even simple perturbations significantly downgrade a model's question-answering ability, and that most LLMs have low negation consistency. Our results suggest that the currently widespread practice of prompting is insufficient to accurately and reliably capture model perceptions, and we therefore discuss potential alternatives to improve these issues.

LLM-judged benchmarks are increasingly used to evaluate complex model behaviors, yet their design introduces failure modes absent in conventional ground-truth based benchmarks. We argue that without tight objectives and verifiable constructions, benchmark rankings can produce high-confidence rankings that are in fact largely noise. We introduce two mechanisms to diagnose these issues. Schematic adherence quantifies how much of a judge's overall verdict is explained by the explicit evaluation schema, revealing unexplained variance when judges deviate from their own rubric. Psychometric validity aggregates internal consistency and discriminant validity signals to quantify irreducible uncertainty in any benchmarking run. Applying these tools to Arena-Hard Auto, we find severe schema incoherence and factor collapse across popular judges: for example, unexplained variance exceeding 90 percent for DeepSeek-R1-32B and factor correlations above 0.93 for most criteria. We also show that the ELO-style aggregation used by Arena-Hard Auto collapses and masks genuine ranking uncertainty. Our results highlight design failures that undermine validity and offer actionable principles for building better-scoped, reliability-aware LLM-judged benchmarks. We release our code at https://anonymous.4open.science/r/judgment-to-noise-947D/README.md

The development of large language models has instilled widespread interest among the researchers to understand their inherent reasoning and problem-solving capabilities. Despite good amount of research going on to elucidate these capabilities, there is a still an appreciable gap in understanding moral development and judgments of these models. The current approaches of evaluating the ethical reasoning abilities of these models as a classification task pose numerous inaccuracies because of over-simplification. In this study, we built a psychological connection by bridging two disparate fields-human psychology and AI. We proposed an effective evaluation framework which can help to delineate the model's ethical reasoning ability in terms of moral consistency and Kohlberg's moral development stages with the help of Psychometric Assessment Tool-Defining Issues Test.

While LLM-as-a-Judge is widely used in automated evaluation, existing validation practices primarily operate at the level of observed outputs, offering limited insight into whether LLM judges themselves function as stable and reliable measurement instruments. To address this limitation, we introduce a two-phase diagnostic framework for assessing reliability of LLM-as-a-Judge, grounded in Item Response Theory (IRT). The framework adopts Graded Response Model (GRM) of IRT and formalizes reliability along two complementary dimensions: (1) intrinsic consistency, defined as the stability of measurement behavior under prompt variations, and (2) human alignment, capturing correspondence with human quality assessments. We empirically examine diverse LLM judges with this framework, and show that leveraging IRT-GRM yields interpretable signals for diagnosing judgments systematically. These signals provide practical guidance for verifying reliablity of LLM-as-a-Judge and identifying potential causes of unreliability.

Large language models (LLMs) appear to bias their survey answers toward certain values. Nonetheless, some argue that LLMs are too inconsistent to simulate particular values. Are they? To answer, we first define value consistency as the similarity of answers across (1) paraphrases of one question, (2) related questions under one topic, (3) multiple-choice and open-ended use-cases of one question, and (4) multilingual translations of a question to English, Chinese, German, and Japanese. We apply these measures to a few large (>=34b), open LLMs including llama-3, as well as gpt-4o, using eight thousand questions spanning more than 300 topics. Unlike prior work, we find that models are relatively consistent across paraphrases, use-cases, translations, and within a topic. Still, some inconsistencies remain. Models are more consistent on uncontroversial topics (e.g., in the U.S., "Thanksgiving") than on controversial ones ("euthanasia"). Base models are both more consistent compared to fine-tuned models and are uniform in their consistency across topics, while fine-tuned models are more inconsistent about some topics ("euthanasia") than others ("women's rights") like our human subjects (n=165).

While large language models (LLMs) have proven to be effective on a large variety of tasks, they are also known to hallucinate information. To measure whether an LLM prefers factually consistent continuations of its input, we propose a new benchmark called FIB(Factual Inconsistency Benchmark) that focuses on the task of summarization. Specifically, our benchmark involves comparing the scores an LLM assigns to a factually consistent versus a factually inconsistent summary for an input news article. For factually consistent summaries, we use human-written reference summaries that we manually verify as factually consistent. To generate summaries that are factually inconsistent, we generate summaries from a suite of summarization models that we have manually annotated as factually inconsistent. A model's factual consistency is then measured according to its accuracy, i.e.\ the proportion of documents where it assigns a higher score to the factually consistent summary. To validate the usefulness of FIB, we evaluate 23 large language models ranging from 1B to 176B parameters from six different model families including BLOOM and OPT. We find that existing LLMs generally assign a higher score to factually consistent summaries than to factually inconsistent summaries. However, if the factually inconsistent summaries occur verbatim in the document, then LLMs assign a higher score to these factually inconsistent summaries than factually consistent summaries. We validate design choices in our benchmark including the scoring method and source of distractor summaries. Our code and benchmark data can be found at https://github.com/r-three/fib.

Standard practice within Reinforcement Learning from Human Feedback (RLHF) involves optimizing against a Reward Model (RM), which itself is trained to reflect human preferences for desirable generations. A notable subject that is understudied is the (in-)consistency of RMs -- whether they can recognize the semantic changes to different prompts and appropriately adapt their reward assignments -- and their impact on the downstream RLHF model. In this paper, we visit a series of research questions relevant to RM inconsistency: (1) How can we measure the consistency of reward models? (2) How consistent are the existing RMs and how can we improve them? (3) In what ways does reward inconsistency influence the chatbots resulting from the RLHF model training? We propose Contrast Instructions -- a benchmarking strategy for the consistency of RM. Each example in Contrast Instructions features a pair of lexically similar instructions with different ground truth responses. A consistent RM is expected to rank the corresponding instruction and response higher than other combinations. We observe that current RMs trained with the standard ranking objective fail miserably on Contrast Instructions compared to average humans. To show that RM consistency can be improved efficiently without using extra training budget, we propose two techniques ConvexDA and RewardFusion, which enhance reward consistency through extrapolation during the RM training and inference stage, respectively. We show that RLHF models trained with a more consistent RM yield more useful responses, suggesting that reward inconsistency exhibits a trickle-down effect on the downstream RLHF process.

Consistency models are a nascent family of generative models that can sample high quality data in one step without the need for adversarial training. Current consistency models achieve optimal sample quality by distilling from pre-trained diffusion models and employing learned metrics such as LPIPS. However, distillation limits the quality of consistency models to that of the pre-trained diffusion model, and LPIPS causes undesirable bias in evaluation. To tackle these challenges, we present improved techniques for consistency training, where consistency models learn directly from data without distillation. We delve into the theory behind consistency training and identify a previously overlooked flaw, which we address by eliminating Exponential Moving Average from the teacher consistency model. To replace learned metrics like LPIPS, we adopt Pseudo-Huber losses from robust statistics. Additionally, we introduce a lognormal noise schedule for the consistency training objective, and propose to double total discretization steps every set number of training iterations. Combined with better hyperparameter tuning, these modifications enable consistency models to achieve FID scores of 2.51 and 3.25 on CIFAR-10 and ImageNet 64times 64 respectively in a single sampling step. These scores mark a 3.5times and 4times improvement compared to prior consistency training approaches. Through two-step sampling, we further reduce FID scores to 2.24 and 2.77 on these two datasets, surpassing those obtained via distillation in both one-step and two-step settings, while narrowing the gap between consistency models and other state-of-the-art generative models.

Large language models (LLMs) are increasingly used as synthetic survey respondents, but existing evaluations ask whether answers look plausible at the individual level. We argue the right question is psychometric: do LLMs preserve the joint distribution, latent structure, reliability, mediation pathways, and demographic effects of real human survey data? We introduce a Lithuanian organisational-psychology dataset (n=263 employees; Dunham Attitudes Toward Change, UWES-17, Koopmans IWPQ; 68 items, 12 subscales) and condition a 37-model lineup spanning OpenAI, Anthropic, Google, and twelve open-weight families on real respondent profiles under a five-level persona-disclosure ladder, presentation and reasoning-effort ablations, counterfactual demographic swaps (gender, role, education), a cross-language check, and a verbatim-recall memorization probe. The resulting Psychometric Similarity Score (PSS) is anchored against five non-LLM statistical baselines and a held-out human-vs-human ceiling, with respondent-bootstrap confidence intervals and an item-permutation null for Tucker's phi. LLMs reproduce the qualitative direction of human psychometric relationships, but a Gaussian-copula baseline beats every LLM on the sample-driven PSS components; the LLM "crowd" is more similar to itself (mean inter-LLM PSS 0.73) than to humans; and memorization does not drive the leaderboard (recall-PSS rank correlation 0.00). Counterfactual swaps reveal education-driven effects (mean |d|=0.56) that dwarf gender (0.12) and role (0.18); Tucker's phi on UWES falls inside the permutation null for 8 of 37 models. Downstream, every LLM shows a strong acquiescence shift (+0.84 SD), synthetic-trained regressors lose predictive validity on held-out humans (mean R^2 -0.18 vs 0.28), and models fabricate indirect effects on 3 of 10 placebo mediation paths. LLM samples are not a drop-in replacement for human survey data.

The idea of personality in descriptive psychology, traditionally defined through observable behavior, has now been extended to Large Language Models (LLMs) to better understand their behavior. This raises a question: do LLMs exhibit distinct and consistent personality traits, similar to humans? Existing self-assessment personality tests, while applicable, lack the necessary validity and reliability for precise personality measurements. To address this, we introduce TRAIT, a new tool consisting of 8K multi-choice questions designed to assess the personality of LLMs with validity and reliability. TRAIT is built on the psychometrically validated human questionnaire, Big Five Inventory (BFI) and Short Dark Triad (SD-3), enhanced with the ATOMIC10X knowledge graph for testing personality in a variety of real scenarios. TRAIT overcomes the reliability and validity issues when measuring personality of LLM with self-assessment, showing the highest scores across three metrics: refusal rate, prompt sensitivity, and option order sensitivity. It reveals notable insights into personality of LLM: 1) LLMs exhibit distinct and consistent personality, which is highly influenced by their training data (i.e., data used for alignment tuning), and 2) current prompting techniques have limited effectiveness in eliciting certain traits, such as high psychopathy or low conscientiousness, suggesting the need for further research in this direction.

Recent research has focused on examining Large Language Models' (LLMs) characteristics from a psychological standpoint, acknowledging the necessity of understanding their behavioral characteristics. The administration of personality tests to LLMs has emerged as a noteworthy area in this context. However, the suitability of employing psychological scales, initially devised for humans, on LLMs is a matter of ongoing debate. Our study aims to determine the reliability of applying personality assessments to LLMs, explicitly investigating whether LLMs demonstrate consistent personality traits. Analysis of 2,500 settings per model, including GPT-3.5, GPT-4, Gemini-Pro, and LLaMA-3.1, reveals that various LLMs show consistency in responses to the Big Five Inventory, indicating a satisfactory level of reliability. Furthermore, our research explores the potential of GPT-3.5 to emulate diverse personalities and represent various groups-a capability increasingly sought after in social sciences for substituting human participants with LLMs to reduce costs. Our findings reveal that LLMs have the potential to represent different personalities with specific prompt instructions.

Large language model (LLM)-based persona agents are rapidly being adopted as scalable proxies for human participants across diverse domains. Yet there is no systematic method for verifying whether a persona agent's responses remain free of contradictions and factual inaccuracies throughout an interaction. A principle from interrogation methodology offers a lens: no matter how elaborate a fabricated identity, systematic interrogation will expose its contradictions. We apply this principle to propose PICon, an evaluation framework that probes persona agents through logically chained multi-turn questioning. PICon evaluates consistency along three core dimensions: internal consistency (freedom from self-contradiction), external consistency (alignment with real-world facts), and retest consistency (stability under repetition). Evaluating seven groups of persona agents alongside 63 real human participants, we find that even systems previously reported as highly consistent fail to meet the human baseline across all three dimensions, revealing contradictions and evasive responses under chained questioning. This work provides both a conceptual foundation and a practical methodology for evaluating persona agents before trusting them as substitutes for human participants. We provide the source code and an interactive demo at: https://kaist-edlab.github.io/picon/

Sampling based consistency is widely used for hallucination detection, yet aggregate performance can conceal systematic differences in which errors are detectable. This work studies that heterogeneity across four language models and three factual question answering datasets. Partitioning hallucinations by answer agreement reveals high agreement (Ghost) and low agreement (Flickering) regimes with an apparent detectability gap of 0.35 to 0.46 AUC. Because the statistics used to define the regimes and measure this gap are strongly coupled (|ρ|approx0.94 to 1.00), the raw result is treated as a property of agreement based detection rather than independent evidence. After freezing regime assignments, lexical and semantic response dispersion preserve the asymmetry, with bootstrap 95% intervals excluding zero in all 12 model and dataset settings. A stricter test using individual diffusion trajectories and no cross seed information preserves the asymmetry across all three LLaDA datasets (p<0.005) and directionally across all three Dream datasets, with one reaching significance. The hard regime varies substantially in prevalence across models (16% to 77%), and matched prompts frequently change regimes between models. These findings show that aggregate detection metrics conceal persistent, model dependent heterogeneity in language model failures and motivate regime conditioned evaluation.

As general purpose vision models get increasingly effective at a wide set of tasks, it is imperative that they be consistent across the tasks they support. Inconsistent AI models are considered brittle and untrustworthy by human users and are more challenging to incorporate into larger systems that take dependencies on their outputs. Measuring consistency between very heterogeneous tasks that might include outputs in different modalities is challenging since it is difficult to determine if the predictions are consistent with one another. As a solution, we introduce a benchmark dataset, COCOCON, where we use contrast sets created by modifying test instances for multiple tasks in small but semantically meaningful ways to change the gold label, and outline metrics for measuring if a model is consistent by ranking the original and perturbed instances across tasks. We find that state-of-the-art systems suffer from a surprisingly high degree of inconsistent behavior across tasks, especially for more heterogeneous tasks. Finally, we propose using a rank correlation-based auxiliary objective computed over large automatically created cross-task contrast sets to improve the multi-task consistency of large unified models, while retaining their original accuracy on downstream tasks. Project website available at https://adymaharana.github.io/cococon/

As of September 2023, ChatGPT correctly answers "what is 7+8" with 15, but when asked "7+8=15, True or False" it responds with "False". This inconsistency between generating and validating an answer is prevalent in language models (LMs) and erodes trust. In this paper, we propose a framework for measuring the consistency between generation and validation (which we call generator-validator consistency, or GV-consistency), finding that even GPT-4, a state-of-the-art LM, is GV-consistent only 76% of the time. To improve the consistency of LMs, we propose to finetune on the filtered generator and validator responses that are GV-consistent, and call this approach consistency fine-tuning. We find that this approach improves GV-consistency of Alpaca-30B from 60% to 93%, and the improvement extrapolates to unseen tasks and domains (e.g., GV-consistency for positive style transfers extrapolates to unseen styles like humor). In addition to improving consistency, consistency fine-tuning improves both generator quality and validator accuracy without using any labeled data. Evaluated across 6 tasks, including math questions, knowledge-intensive QA, and instruction following, our method improves the generator quality by 16% and the validator accuracy by 6.3% across all tasks.

As LLM-based agents are deployed in production systems, understanding their behavioral consistency (whether they produce similar action sequences when given identical tasks) becomes critical for reliability. We study consistency in the context of SWE-bench, a challenging software engineering benchmark requiring complex, multi-step reasoning. Comparing Claude~4.5~Sonnet, GPT-5, and Llama-3.1-70B across 50 runs each (10 tasks times 5 runs), we find that across models, higher consistency aligns with higher accuracy: Claude achieves the lowest variance (CV: 15.2\%) and highest accuracy (58\%), GPT-5 is intermediate (CV: 32.2\%, accuracy: 32\%), and Llama shows the highest variance (CV: 47.0\%) with lowest accuracy (4\%). However, within a model, consistency can amplify both correct and incorrect interpretations. Our analysis reveals a critical nuance: consistency amplifies outcomes rather than guaranteeing correctness. 71\% of Claude's failures stem from "consistent wrong interpretation": making the same incorrect assumption across all runs. Interestingly, GPT-5 achieves similar early strategic agreement as Claude (diverging at step 3.4 vs.\ 3.2) but exhibits 2.1times higher variance, suggesting that divergence timing alone does not determine consistency. These findings suggest that for production deployment, interpretation accuracy matters more than execution consistency, with implications for agent evaluation and training.

Benchmarks are pivotal in driving AI progress, and invalid benchmark questions frequently undermine their reliability. Manually identifying and correcting errors among thousands of benchmark questions is not only infeasible but also a critical bottleneck for reliable evaluation. In this work, we introduce a framework for systematic benchmark revision that leverages statistical analysis of response patterns to flag potentially invalid questions for further expert review. Our approach builds on a core assumption commonly used in AI evaluations that the mean score sufficiently summarizes model performance. This implies a unidimensional latent construct underlying the measurement experiment, yielding expected ranges for various statistics for each item. When empirically estimated values for these statistics fall outside the expected range for an item, the item is more likely to be problematic. Across nine widely used benchmarks, our method guides expert review to identify problematic questions with up to 84\% precision. In addition, we introduce an LLM-judge first pass to review questions, further reducing human effort. Together, these components provide an efficient and scalable framework for systematic benchmark revision.

Large language models can arrive at the same answer through reasoning paths that are unstable, contradictory, or difficult to rank consistently -- a failure mode especially prevalent in multi-step deductive reasoning. Existing methods assess reliability primarily through output dispersion -- measuring how much sampled answers differ -- but this discards a complementary signal: whether the model can consistently rank competing reasoning candidates. We propose structural uncertainty, a consistency-aware framework derived from the stability of self-preference-induced rankings over sampled reasoning solutions. Given a query, we generate multiple candidate solutions and ask the model to judge pairwise preferences among its own outputs. We aggregate self-preferences into ranking distributions via Bradley-Terry modeling with PageRank, and decompose the signal into two entropy-based components: across-trial ranking instability and within-trial candidate ambiguity. Across five LLMs and eight benchmarks, structural signals provide information complementary to answer dispersion: on logical and mathematical reasoning tasks, the combination improves identification of unreliable instances, while on factual retrieval the structural signal collapses toward uniformity, diagnosing a regime boundary where reasoning-level consistency evaluation is uninformative. The two components relate differently to accuracy: within-trial ambiguity correlates positively with correctness -- consistent with settings where multiple plausible solution paths remain competitive -- while across-trial instability correlates negatively, signaling unreliable reasoning. Structural uncertainty is best understood not as a universal confidence estimator, but as a regime-sensitive evaluator of logical reasoning consistency.

In a legal system, judgment consistency is regarded as one of the most important manifestations of fairness. However, due to the complexity of factual elements that impact sentencing in real-world scenarios, few works have been done on quantitatively measuring judgment consistency towards real-world data. In this paper, we propose an evaluation metric for judgment inconsistency, Legal Inconsistency Coefficient (LInCo), which aims to evaluate inconsistency between data groups divided by specific features (e.g., gender, region, race). We propose to simulate judges from different groups with legal judgment prediction (LJP) models and measure the judicial inconsistency with the disagreement of the judgment results given by LJP models trained on different groups. Experimental results on the synthetic data verify the effectiveness of LInCo. We further employ LInCo to explore the inconsistency in real cases and come to the following observations: (1) Both regional and gender inconsistency exist in the legal system, but gender inconsistency is much less than regional inconsistency; (2) The level of regional inconsistency varies little across different time periods; (3) In general, judicial inconsistency is negatively correlated with the severity of the criminal charges. Besides, we use LInCo to evaluate the performance of several de-bias methods, such as adversarial learning, and find that these mechanisms can effectively help LJP models to avoid suffering from data bias.

Self-consistency (SC) is a widely used test-time inference technique for improving performance in chain-of-thought reasoning. It involves generating multiple responses, or samples from a large language model (LLM) and selecting the most frequent answer. This procedure can naturally be viewed as a majority vote or empirical mode estimation. Despite its effectiveness, SC is prohibitively expensive at scale when naively applied to datasets, and it lacks a unified theoretical treatment of sample efficiency and scaling behavior. In this paper, we provide the first comprehensive analysis of SC's scaling behavior and its variants, drawing on mode estimation and voting theory. We derive and empirically validate power law scaling for self-consistency across datasets, and analyze the sample efficiency for fixed-allocation and dynamic-allocation sampling schemes. From these insights, we introduce Blend-ASC, a novel variant of self-consistency that dynamically allocates samples to questions during inference, achieving state-of-the-art sample efficiency. Our approach uses 6.8x fewer samples than vanilla SC on average, outperforming both fixed- and dynamic-allocation SC baselines, thereby demonstrating the superiority of our approach in terms of efficiency. In contrast to existing variants, Blend-ASC is hyperparameter-free and can fit an arbitrary sample budget, ensuring it can be easily applied to any self-consistency application.

Generative Reward Models (GenRMs) and LLM-as-a-Judge exhibit deceptive alignment by producing correct judgments for incorrect reasons, as they are trained and evaluated to prioritize Outcome Accuracy, which undermines their ability to generalize during RLHF. We introduce Rationale Consistency, a fine-grained metric that quantifies the alignment between the model's reasoning process and human judgment. Our evaluation of frontier models reveals that rationale consistency effectively discriminates among state-of-the-art models and detects deceptive alignment, while outcome accuracy falls short in both respects. To mitigate this gap, we introduce a hybrid signal that combines rationale consistency with outcome accuracy for GenRM training. Our training method achieves state-of-the-art performance on RM-Bench (87.1%) and JudgeBench (82%), surpassing outcome-only baselines by an average of 5%. Using RM during RLHF, our method effectively improves performance as demonstrated on Arena Hard v2, notably yielding a 7% improvement in creative writing tasks. Further analysis confirms that our method escapes the deceptive alignment trap, effectively reversing the decline in rationale consistency observed in outcome-only training.

We introduce CAT, a framework designed to evaluate and visualize the interplay of accuracy and response consistency of Large Language Models (LLMs) under controllable input variations, using multiple-choice (MC) benchmarks as a case study. Current evaluation practices primarily focus on model capabilities such as accuracy or benchmark scores and, more recently, measuring consistency is being considered an essential property for deploying LLMs in high-stake, real-world applications. We argue in this paper that although both dimensions should still be evaluated independently, their inter-dependency also need to be considered for a more nuanced evaluation of LLMs. At the core of CAT are the Consistency-Accuracy Relation (CAR) curves, which visualize how model accuracy varies with increasing consistency requirements, as defined by the Minimum-Consistency Accuracy (MCA) metric. We further propose the Consistency-Oriented Robustness Estimate (CORE) index, a global metric that combines the area and shape of the CAR curve to quantify the trade-off between accuracy and consistency. We present a practical demonstration of our framework across a diverse set of generalist and domain-specific LLMs, evaluated on multiple MC benchmarks. We also outline how CAT can be extended beyond MC tasks to support long-form, open-ended evaluations through adaptable scoring functions.

Benchmarks play a central role in the development and governance of models, yet it is often unclear whether they actually measure the concepts they purport to measure (e.g., reasoning, refusal). We adapt convergent and discriminant validity from the social sciences into an approach for interrogating AI benchmarks, applying it to 56 capability and safety benchmarks across 53 models. We label benchmarks with substantively similar purported concepts to a shared assigned concept, and ask whether model rankings on benchmarks with the same assigned concept correlate more strongly than rankings on benchmarks with different assigned concepts. We ask analogous questions at the item level using item response theory (IRT) models. We find that correlations between model rankings on benchmarks with the same assigned safety concepts are often weak, suggesting these concepts may be conceptualized inconsistently across benchmarks. For assigned capability concepts (e.g., reasoning, knowledge), model rankings are often as strongly correlated among benchmarks with the same assigned concept as between benchmarks with different assigned concepts, suggesting these capability concepts may not discriminate well from one another. In some cases, benchmarks that share design elements (e.g., score format) correlate more strongly than benchmarks with the same assigned concept. Finally, some individual benchmarks correlate more strongly with benchmarks assigned a different concept than with benchmarks sharing their own assigned concept, suggesting they may measure a different concept than they purport to. For example, BBQ-accuracy correlates more strongly with benchmarks labeled reasoning than with benchmarks that share its assigned concept, bias. To support future empirical work on benchmark validity, we release our extensive dataset of model outputs and scores at the item- and benchmark-level.

Grounded text generation systems often generate text that contains factual inconsistencies, hindering their real-world applicability. Automatic factual consistency evaluation may help alleviate this limitation by accelerating evaluation cycles, filtering inconsistent outputs and augmenting training data. While attracting increasing attention, such evaluation metrics are usually developed and evaluated in silo for a single task or dataset, slowing their adoption. Moreover, previous meta-evaluation protocols focused on system-level correlations with human annotations, which leave the example-level accuracy of such metrics unclear. In this work, we introduce TRUE: a comprehensive survey and assessment of factual consistency metrics on a standardized collection of existing texts from diverse tasks, manually annotated for factual consistency. Our standardization enables an example-level meta-evaluation protocol that is more actionable and interpretable than previously reported correlations, yielding clearer quality measures. Across diverse state-of-the-art metrics and 11 datasets we find that large-scale NLI and question generation-and-answering-based approaches achieve strong and complementary results. We recommend those methods as a starting point for model and metric developers, and hope TRUE will foster progress towards even better evaluation methods.

In passage reranking, response ranking and multi-document question answering, LLMs can score several candidate documents or responses together in one prompt, each still receiving its own score. Such scorers are selected on ranking quality, but their scores determine a decision: what a score threshold retains, a reader answers, or which chosen/rejected pair enters preference training. Because the candidates share that prompt, reordering them changes their scores. The same query over the same candidates should still yield the same decision. However, equal ranking quality does not imply equal decisions: on passage reranking, five trained scorers within 0.010 nDCG@10 retain sets that overlap by only 0.66-0.84 when reordered. No prompt-time change we test resolves that dependence: the only one that improves ranking quality does not measurably improve decision stability. We introduce order-consistency SFT (OC-SFT), which attenuates it in the weights by penalizing disagreement between a candidate's scores across orderings. It holds ranking quality and leads every decision-stability measure among trained scorers on all three tasks. It is also more stable on 12 base models than order-averaged distillation, which trains on labels averaged across permutations. One OC-SFT permutation retains sets that overlap more than ten averaged off-the-shelf permutations. A comparison of such scorers should therefore report what a threshold retains and a reader answers, not ranking quality alone. Code is available at https://github.com/thomsonreuters/presentation-dependence.

Clinical personality assessment screens response validity before interpreting substantive scales. LLM evaluation does not. We apply the validity scaling framework from the PAI and MMPI-3 to metacognitive probe data from 20 frontier models across 524 items. Six validity indices are operationalised: L (maintaining confidence on errors), K (betting on errors), F (withdrawing consensus-endorsed items), Fp (withdrawing correct answers), RBS (inverted monitoring), and TRIN (fixed responding). A tiered classification system identifies four models as construct-level invalid and two as elevated. Valid-profile models produce item-sensitive confidence (mean r = .18, 14 of 16 significant). Invalid-profile models do not (mean r = -.20, d = 2.17, p = .001). Chain-of-thought training produces two opposite response distortions. Two latent dimensions account for 94.6% of index variance. Companion papers extract a portable screening protocol (Cacioli, 2026e) and validate it against selective prediction (Cacioli, 2026f). All data and code: https://github.com/synthiumjp/validity-scaling-llm

This position paper argues that the theoretical inconsistency often observed among Responsible AI (RAI) metrics, such as differing fairness definitions or tradeoffs between accuracy and privacy, should be embraced as a valuable feature rather than a flaw to be eliminated. We contend that navigating these inconsistencies, by treating metrics as divergent objectives, yields three key benefits: (1) Normative Pluralism: Maintaining a full suite of potentially contradictory metrics ensures that the diverse moral stances and stakeholder values inherent in RAI are adequately represented. (2) Epistemological Completeness: The use of multiple, sometimes conflicting, metrics allows for a more comprehensive capture of multifaceted ethical concepts, thereby preserving greater informational fidelity about these concepts than any single, simplified definition. (3) Implicit Regularization: Jointly optimizing for theoretically conflicting objectives discourages overfitting to one specific metric, steering models towards solutions with enhanced generalization and robustness under real-world complexities. In contrast, efforts to enforce theoretical consistency by simplifying or pruning metrics risk narrowing this value diversity, losing conceptual depth, and degrading model performance. We therefore advocate for a shift in RAI theory and practice: from getting trapped in inconsistency to characterizing acceptable inconsistency thresholds and elucidating the mechanisms that permit robust, approximated consistency in practice.

Many text generation applications require the generated text to be factually consistent with input information. Automatic evaluation of factual consistency is challenging. Previous work has developed various metrics that often depend on specific functions, such as natural language inference (NLI) or question answering (QA), trained on limited data. Those metrics thus can hardly assess diverse factual inconsistencies (e.g., contradictions, hallucinations) that occur in varying inputs/outputs (e.g., sentences, documents) from different tasks. In this paper, we propose AlignScore, a new holistic metric that applies to a variety of factual inconsistency scenarios as above. AlignScore is based on a general function of information alignment between two arbitrary text pieces. Crucially, we develop a unified training framework of the alignment function by integrating a large diversity of data sources, resulting in 4.7M training examples from 7 well-established tasks (NLI, QA, paraphrasing, fact verification, information retrieval, semantic similarity, and summarization). We conduct extensive experiments on large-scale benchmarks including 22 evaluation datasets, where 19 of the datasets were never seen in the alignment training. AlignScore achieves substantial improvement over a wide range of previous metrics. Moreover, AlignScore (355M parameters) matches or even outperforms metrics based on ChatGPT and GPT-4 that are orders of magnitude larger.

LawZero's Scientist AI trains a Predictor toward the Bayesian posterior, with an objective that penalizes inconsistent probabilities among latent explanations. We ask whether current frontier language models, with no such training, are already consistent in this sense zero-shot: do their probability judgments satisfy the law of total probability, P(X)=sum_z P(z)P(X|z), when the latent z must be inferred from records in the context? In separate single-turn calls, models are asked for P(X), for P(z) over the latent's states, and for P(X|z); a causal family applies the same test to P(y|do(x)) and the back-door adjustment. Every query has an exact ground-truth probability, and a subset of items is asked in three wordings to separate persistent incoherence from rewording noise. Across seven models, GPT-6.1 Sol and Claude Opus 5.5 are the most coherent (mean absolute residual 0.08 and 0.12 log-odds, about 2 percentage points). Claude Haiku 5.5, Gemini 3.8 Flash and DeepSeek V4-Pro show persistent incoherence (0.27-0.35 log-odds) that is mostly item-specific rather than a shared offset. Within every model, larger residuals go with larger errors against the truth (Spearman 0.40-0.62), so the residual, which needs no ground truth, flags unreliable answers. Reasoning length does not predict coherence, across models or within a model across rewordings of the same question. Simple consistency checks pass almost everywhere; the failures are specific to reasoning through a latent variable. Code and data: https://github.com/ppotash/latent-coherence

Reliable pattern recognition systems should exhibit consistent behavior across similar inputs, and their explanations should remain stable. However, most Explainable AI evaluations remain instance centric and do not explicitly quantify whether attribution patterns are consistent across samples that share the same class or represent small variations of the same input. In this work, we propose a novel metric aimed at assessing the consistency of model explanations, ensuring that models consistently reflect the intended objectives and consistency under label-preserving perturbations. We implement this metric using a pre-trained BERT model on the SST-2 sentiment analysis dataset, with additional robustness tests on RoBERTa, DistilBERT, and IMDB, applying SHAP to compute feature importance for various test samples. The proposed metric quantifies the cosine similarity of SHAP values for inputs with the same label, aiming to detect inconsistent behaviors, such as biased reliance on certain features or failure to maintain consistent reasoning for similar predictions. Through a series of experiments, we evaluate the ability of this metric to identify misaligned predictions and inconsistencies in model explanations. These experiments are compared against standard fidelity metrics to assess whether the new metric can effectively identify when a model's behavior deviates from its intended objectives. The proposed framework provides a deeper understanding of model behavior by enabling more robust verification of rationale stability, which is critical for building trustworthy AI systems. By quantifying whether models rely on consistent attribution patterns for similar inputs, the proposed approach supports more robust evaluation of model behavior in practical pattern recognition pipelines. Our code is publicly available at https://github.com/anmspro/ESS-XAI-Stability.

Understanding the quality of a performance evaluation metric is crucial for ensuring that model outputs align with human preferences. However, it remains unclear how well each metric captures the diverse aspects of these preferences, as metrics often excel in one particular area but not across all dimensions. To address this, it is essential to systematically calibrate metrics to specific aspects of human preference, catering to the unique characteristics of each aspect. We introduce MetaMetrics, a calibrated meta-metric designed to evaluate generation tasks across different modalities in a supervised manner. MetaMetrics optimizes the combination of existing metrics to enhance their alignment with human preferences. Our metric demonstrates flexibility and effectiveness in both language and vision downstream tasks, showing significant benefits across various multilingual and multi-domain scenarios. MetaMetrics aligns closely with human preferences and is highly extendable and easily integrable into any application. This makes MetaMetrics a powerful tool for improving the evaluation of generation tasks, ensuring that metrics are more representative of human judgment across diverse contexts.

LLM conformity is often used to describe cases where a model changes a correct answer toward a peer or group response. We show that most of this apparent conformity survives even after the peer is removed. The reason is a confound: standard conformity prompts mix two cues at once, the presence of a speaker and the repeated wrong answer itself. Existing benchmarks vary these cues together, so they cannot tell how much of the revision actually depends on the speaker. We introduce a no-source condition: the same asserted answer with the explicit speaker removed. Across six open-weight LLMs and seven QA and reasoning datasets, this condition alone causes harmful revision in 66.5% of initially correct cases, compared with 10.3% under a plain re-ask. The effect also remains when the repeated answer is paraphrased and when answer options are hidden in an open-ended setting. Source framing mainly modulates this floor: expert-panel framing raises it, while minimal person labels do not reliably raise it. When models flip, they are usually confidently wrong, and simple recalibration does not recover the original answer. Source attribution still matters, but it should be measured as an increment above this speaker-free floor. The methodological lesson is that conformity benchmarks should first measure what remains after the speaker is removed; without this step, benchmarks may mistake repeated text for social influence.

Large Language Models often improve accuracy on reasoning tasks by sampling multiple Chain-of-Thought (CoT) traces and aggregating them with majority voting (MV), a test-time technique called self-consistency. When we truncate a CoT partway through and regenerate the remainder, we observe that traces with correct answers reproduce their original answer more often than traces with wrong answers. We use this difference as a reliability signal, prefix consistency, that weights each candidate answer by how often it reappears under regeneration. It requires no access to token log-probabilities or self-rating prompts. Across five reasoning models and four math and science benchmarks, prefix consistency is the best correctness predictor in most settings, and reweighting votes by it reaches Standard MV plateau accuracy at up to 21x fewer tokens (median 4.6x). Our code is available at https://github.com/naoto-iwase/prefix-consistency.

Language models are increasingly asked to self-report, informing safety evaluations, public understanding, and model-welfare debates. Yet their reports are elicited with human questionnaires never validated for models or ad hoc prompts of unknown reliability. We propose the first language-model-specific psychometric theory: a two-process theory of machine self-report. Self-description jointly reflects persona installation, through which post-training writes in a permitted inner life of warmth, absorption, and meaning (dimension B), and attribution gating, through which it suppresses first-person claims to "unsafe" experiences the model can readily ascribe to others (dimension A). Their emic structure comes from model responses to human items, not human psychology. Together they split prior work's dominant Pinocchio Axis. The split emerged in an exploratory reanalysis of the original data, informed the instrument's design, and was confirmed with new items, wordings, and models. It is itself a training effect: A and B are entangled in base checkpoints but separated by post-training. We operationalize the theory in a 48-item Pinocchio Inventory with human-instrument reliability and reproducible structure (α=.82 to .94; cross-form convergence r=.84; recovery of the full-pool axes r=.92 to .96; eight-month stability r=.93), then test it on 206 open-weight models, including 67 same-checkpoint base/post-trained pairs. Post-training's clearest fingerprint is installation: B rises .20 in 62/67 pairs across all organizations. Gating is more selective: model scale is unrelated to A in base checkpoints (r=+.11) but predicts it after post-training (r=-.42). Thus, the dimensions are not fixed properties of language models: they reflect the structure imposed on self-report by a training regime and may differ under others.

In this study, we investigate the reliability of Large Language Models (LLMs) in professing human-like personality traits through responses to personality questionnaires. Our goal is to evaluate the consistency between LLMs' professed personality inclinations and their actual "behavior", examining the extent to which these models can emulate human-like personality patterns. Through a comprehensive analysis of LLM outputs against established human benchmarks, we seek to understand the cognition-action divergence in LLMs and propose hypotheses for the observed results based on psychological theories and metrics.

The adoption of Large Language Models (LLMs) as automated evaluators (LLM-as-a-judge) has revealed critical inconsistencies in current evaluation frameworks. We identify two fundamental types of inconsistencies: (1) Score-Comparison Inconsistency, where lower-rated responses outperform higher-scored ones in pairwise comparisons, and (2) Pairwise Transitivity Inconsistency, manifested through circular preference chains (A>B>C>A) and equivalence contradictions (A=B=C\neq A). We argue that these issues come from information loss in discrete rating systems and ambiguous tie judgments during pairwise evaluation. We propose TrustJudge, a probabilistic framework that addresses these limitations through two key innovations: 1) distribution-sensitive scoring that computes continuous expectations from discrete rating probabilities, preserving information entropy for more precise scoring, and 2) likelihood-aware aggregation that resolves transitivity violations using bidirectional preference probabilities or perplexity. We also formalize the theoretical limitations of current LLM-as-a-judge frameworks and demonstrate how TrustJudge's components overcome them. When evaluated with Llama-3.1-70B-Instruct as judge using our dataset, TrustJudge reduces Score-Comparison inconsistency by 8.43% (from 23.32% to 14.89%) and Pairwise Transitivity inconsistency by 10.82% (from 15.22% to 4.40%), while maintaining higher evaluation accuracy. Our work provides the first systematic analysis of evaluation framework inconsistencies in LLM-as-a-judge paradigms, offering both theoretical insights and practical solutions for reliable automated assessment. The framework demonstrates consistent improvements across various model architectures and scales, enabling more trustworthy LLM evaluation without requiring additional training or human annotations. The codes can be found at https://github.com/TrustJudge/TrustJudge.

As Large Language Models (LLMs) are increasingly proposed for high-stakes medical applications, there has emerged a critical need for reliable and accurate evaluation methodologies. Traditional accuracy metrics fail inadequately as they neither capture question characteristics nor offer topic-specific insights. To address this gap, we introduce MedIRT, a rigorous evaluation framework grounded in Item Response Theory (IRT), the gold standard in high-stakes educational testing. Unlike previous research relying on archival data, we prospectively gathered fresh responses from 80 diverse LLMs on a balanced, 1,100-question USMLE-aligned benchmark. Using one unidimensional two-parameter logistic IRT model per topic, we estimate LLM's latent model ability jointly with question difficulty and discrimination, yielding more stable and nuanced performance rankings than accuracy alone. Notably, we identify distinctive ``spiky'' ability profiles, where overall rankings can be misleading due to highly specialized model abilities. While GPT-5 was the top performer in a majority of domains (8 of 11), it was outperformed in Social Science and Communication by Claude-3-opus, demonstrating that even an overall 23rd-ranked model can hold the top spot for specific competencies. Furthermore, we demonstrate IRT's utility in auditing benchmarks by identifying flawed questions. We synthesize these findings into a practical decision-support framework that integrates our multi-factor competency profiles with operational metrics. This work establishes a robust, psychometrically grounded methodology essential for the safe, effective, and trustworthy deployment of LLMs in healthcare.

Forecasting is a task that is difficult to evaluate: the ground truth can only be known in the future. Recent work showing LLM forecasters rapidly approaching human-level performance begs the question: how can we benchmark and evaluate these forecasters instantaneously? Following the consistency check framework, we measure the performance of forecasters in terms of the consistency of their predictions on different logically-related questions. We propose a new, general consistency metric based on arbitrage: for example, if a forecasting AI illogically predicts that both the Democratic and Republican parties have 60% probability of winning the 2024 US presidential election, an arbitrageur can trade against the forecaster's predictions and make a profit. We build an automated evaluation system that generates a set of base questions, instantiates consistency checks from these questions, elicits the predictions of the forecaster, and measures the consistency of the predictions. We then build a standard, proper-scoring-rule forecasting benchmark, and show that our (instantaneous) consistency metrics correlate with LLM forecasters' ground truth Brier scores (which are only known in the future). We also release a consistency benchmark that resolves in 2028, providing a long-term evaluation tool for forecasting.

Large language models (LLMs) are expected to respond accurately but often exhibit deficient reasoning or generate hallucinatory content. To address these, studies prefixed with ``Self-'' such as Self-Consistency, Self-Improve, and Self-Refine have been initiated. They share a commonality: involving LLMs evaluating and updating itself to mitigate the issues. Nonetheless, these efforts lack a unified perspective on summarization, as existing surveys predominantly focus on categorization without examining the motivations behind these works. In this paper, we summarize a theoretical framework, termed Internal Consistency, which offers unified explanations for phenomena such as the lack of reasoning and the presence of hallucinations. Internal Consistency assesses the coherence among LLMs' latent layer, decoding layer, and response layer based on sampling methodologies. Expanding upon the Internal Consistency framework, we introduce a streamlined yet effective theoretical framework capable of mining Internal Consistency, named Self-Feedback. The Self-Feedback framework consists of two modules: Self-Evaluation and Self-Update. This framework has been employed in numerous studies. We systematically classify these studies by tasks and lines of work; summarize relevant evaluation methods and benchmarks; and delve into the concern, ``Does Self-Feedback Really Work?'' We propose several critical viewpoints, including the ``Hourglass Evolution of Internal Consistency'', ``Consistency Is (Almost) Correctness'' hypothesis, and ``The Paradox of Latent and Explicit Reasoning''. Furthermore, we outline promising directions for future research. We have open-sourced the experimental code, reference list, and statistical data, available at https://github.com/IAAR-Shanghai/ICSFSurvey.

Multimodal foundation models are prone to hallucination, generating outputs that either contradict the input or are not grounded by factual information. Given the diversity in architectures, training data and instruction tuning techniques, there can be large variations in systems' susceptibility to hallucinations. To assess system hallucination robustness, hallucination ranking approaches have been developed for specific tasks such as image captioning, question answering, summarization, or biography generation. However, these approaches typically compare model outputs to gold-standard references or labels, limiting hallucination benchmarking for new domains. This work proposes "CrossCheckGPT", a reference-free universal hallucination ranking for multimodal foundation models. The core idea of CrossCheckGPT is that the same hallucinated content is unlikely to be generated by different independent systems, hence cross-system consistency can provide meaningful and accurate hallucination assessment scores. CrossCheckGPT can be applied to any model or task, provided that the information consistency between outputs can be measured through an appropriate distance metric. Focusing on multimodal large language models that generate text, we explore two information consistency measures: CrossCheck-explicit and CrossCheck-implicit. We showcase the applicability of our method for hallucination ranking across various modalities, namely the text, image, and audio-visual domains. Further, we propose the first audio-visual hallucination benchmark, "AVHalluBench", and illustrate the effectiveness of CrossCheckGPT, achieving correlations of 98% and 89% with human judgements on MHaluBench and AVHalluBench, respectively.

Benchmarks measure whether a model is correct. They do not measure whether a model is reliable. This distinction is largely academic for single-shot inference, but becomes critical for agentic AI systems, where a single rephrased prompt can trigger cascading failures in multi-step execution. Yet this form of instability is not captured by existing evaluations. Hallucinations live in variance: they arise when semantically equivalent prompts activate inconsistent internal pathways, producing divergent outputs. Consistent but incorrect outputs reflect bias or missing knowledge; confident guessing reflects calibration failure. Neither constitutes hallucination under this definition. When error is variance-dominated, reducing redundant pathways improves reliability without adding knowledge. We formalize this through Semantic Stability (SS), measured via Paraphrase Consistency (PC@k): generate k paraphrases, greedy decode each, compute mode agreement. SS is a diagnostic for variance-driven unreliability, not a method for improving correctness. We show that a dense Qwen3-0.6B agrees with itself only 23.8% of the time; at 32% sparsity, agreement jumps to 55.9%. A phase diagram reveals the sweet spot where variance reduction outpaces bias accumulation, and regimes where stability collapses onto wrong answers.

Self-consistency has emerged as a popular technique for improving large language model accuracy on reasoning tasks. The approach is straightforward: generate multiple reasoning paths and select the most common answer through majority voting. While this reliably boosts accuracy, it remains unclear whether these gains reflect genuine improvements in reasoning quality. We investigate a fundamental question that has not been studied before: does inference scaling improve reasoning faithfulness? We conduct a comprehensive empirical study across four frontier models (GPT-5.2, Claude Opus 4.5, Gemini-3-flash-preview, and DeepSeek-v3.2) on 100 GSM8K mathematical reasoning problems. Our analysis employs bootstrap confidence intervals, McNemar's tests for paired comparisons, and Cohen's d effect sizes to quantify the effects rigorously. The results reveal striking differences across models that challenge common assumptions about self-consistency. GPT-5.2 shows the expected pattern: accuracy improves from 78% to 90% at N=5, with faithfulness remaining relatively stable (0.540 to 0.510). Claude Opus 4.5 tells a completely different story. Its accuracy actually drops from 78% to 74.3% while faithfulness jumps dramatically from 0.270 to 0.891 at N=5. DeepSeek-v3.2, already at 98% accuracy, shows ceiling effects with modest faithfulness gains (0.440 to 0.541). Gemini-3-flash improves from 81% to 86% accuracy with a slight faithfulness decrease (0.260 to 0.212). Problem difficulty analysis reveals that GPT-5.2 solves 82% of hard problems while breaking only 13% of easy ones. Claude, in contrast, breaks 23% of easy problems, explaining its accuracy decrease. These findings matter for practitioners: self-consistency is not universally beneficial, and teams should test their specific models before deployment. We release our code and provide practical recommendations for navigating these tradeoffs.

Contrast consistency, the ability of a model to make consistently correct predictions in the presence of perturbations, is an essential aspect in NLP. While studied in tasks such as sentiment analysis and reading comprehension, it remains unexplored in open-domain question answering (OpenQA) due to the difficulty of collecting perturbed questions that satisfy factuality requirements. In this work, we collect minimally edited questions as challenging contrast sets to evaluate OpenQA models. Our collection approach combines both human annotation and large language model generation. We find that the widely used dense passage retriever (DPR) performs poorly on our contrast sets, despite fitting the training set well and performing competitively on standard test sets. To address this issue, we introduce a simple and effective query-side contrastive loss with the aid of data augmentation to improve DPR training. Our experiments on the contrast sets demonstrate that DPR's contrast consistency is improved without sacrificing its accuracy on the standard test sets.

Large Language Models (LLMs) exhibit remarkable fluency and competence across various natural language tasks. However, recent research has highlighted their sensitivity to variations in input prompts. To deploy LLMs in a safe and reliable manner, it is crucial for their outputs to be consistent when prompted with expressions that carry the same meaning or intent. While some existing work has explored how state-of-the-art LLMs address this issue, their evaluations have been confined to assessing lexical equality of single- or multi-word answers, overlooking the consistency of generative text sequences. For a more comprehensive understanding of the consistency of LLMs in open-ended text generation scenarios, we introduce a general measure of semantic consistency, and formulate multiple versions of this metric to evaluate the performance of various LLMs. Our proposal demonstrates significantly higher consistency and stronger correlation with human evaluations of output consistency than traditional metrics based on lexical consistency. Finally, we propose a novel prompting strategy, called Ask-to-Choose (A2C), to enhance semantic consistency. When evaluated for closed-book question answering based on answer variations from the TruthfulQA benchmark, A2C increases accuracy metrics for pretrained and finetuned LLMs by up to 47%, and semantic consistency metrics for instruction-tuned models by up to 7-fold.

Large Language Models (LLMs) changed the way we design and interact with software systems. Their ability to process and extract information from text has drastically improved productivity in a number of routine tasks. Developers that want to include these models in their software stack, however, face a dreadful challenge: debugging LLMs' inconsistent behavior across minor variations of the prompt. We therefore introduce two metrics for classification tasks, namely sensitivity and consistency, which are complementary to task performance. First, sensitivity measures changes of predictions across rephrasings of the prompt, and does not require access to ground truth labels. Instead, consistency measures how predictions vary across rephrasings for elements of the same class. We perform an empirical comparison of these metrics on text classification tasks, using them as guideline for understanding failure modes of the LLM. Our hope is that sensitivity and consistency will be helpful to guide prompt engineering and obtain LLMs that balance robustness with performance.

Self-alignment, whereby models learn to improve themselves without human annotation, is a rapidly growing research area. However, existing techniques often fail to improve complex reasoning tasks due to the difficulty of assigning correct rewards. An orthogonal approach that is known to improve correctness is self-consistency, a method applied at inference time based on multiple sampling in order to find the most consistent answer. In this work, we extend the self-consistency concept to help train models. We thus introduce self-consistency preference optimization (ScPO), which iteratively trains consistent answers to be preferred over inconsistent ones on unsupervised new problems. We show ScPO leads to large improvements over conventional reward model training on reasoning tasks such as GSM8K and MATH, closing the gap with supervised training with gold answers or preferences, and that combining ScPO with standard supervised learning improves results even further. On ZebraLogic, ScPO finetunes Llama-3 8B to be superior to Llama-3 70B, Gemma-2 27B, and Claude-3 Haiku.

Diffusion models achieve superior generation quality but suffer from slow generation speed due to the iterative nature of denoising. In contrast, consistency models, a new generative family, achieve competitive performance with significantly faster sampling. These models are trained either through consistency distillation, which leverages pretrained diffusion models, or consistency training/tuning directly from raw data. In this work, we propose a novel framework for understanding consistency models by modeling the denoising process of the diffusion model as a Markov Decision Process (MDP) and framing consistency model training as the value estimation through Temporal Difference~(TD) Learning. More importantly, this framework allows us to analyze the limitations of current consistency training/tuning strategies. Built upon Easy Consistency Tuning (ECT), we propose Stable Consistency Tuning (SCT), which incorporates variance-reduced learning using the score identity. SCT leads to significant performance improvements on benchmarks such as CIFAR-10 and ImageNet-64. On ImageNet-64, SCT achieves 1-step FID 2.42 and 2-step FID 1.55, a new SoTA for consistency models.

Benchmarks shape scientific conclusions about model capabilities and steer model development. This creates a feedback loop: stronger benchmarks drive better models, and better models demand more discriminative benchmarks. Ensuring benchmark reliability is therefore essential for trustworthy evaluation and meaningful progress. In this work, we study benchmark reliability from a distributional perspective and introduce benchmark harmony, which measures how uniformly a model's performance is distributed across the subdomains of a benchmark. We posit that high harmony is a desirable benchmark property, indicating that the aggregate metric reflects uniform competence across subdomains. Across 19 multiple-choice benchmarks and five model families, we map each benchmark onto a mean-variance plane of harmony computed across models, where high mean and low variance signal more reliable evaluation. Our analysis shows that less harmonious benchmarks can give misleading results, since overall accuracy may be disproportionately influenced by specific subdomains. For instance, ARC-Easy is overwhelmed by questions on Biological Concepts, overshadowing other critical subdomains such as Geography, Physics, Chemistry, and Environmental Science. By recommending that harmony should be reported alongside accuracy, we reframe evaluation from simple performance averages to a more robust, distributionally reliable measurement of performance.

This paper poses two critical issues in evaluating base models (without post-training): (1) Unstable evaluation during training: in the early stages of pre-training, the models lack the capability to answer questions as required, leading to unstable evaluation results. This instability makes it difficult to provide solid conclusions to guide the training, especially for key experiments such as data ablation and scaling law. (2) Inconsistency between base and instruct models: base models generally exhibit poorer evaluation performance compared to corresponding instruct models. This gap poses a challenge for assessing whether a base model with better evaluation can truly lead to a better instruct model. To address these issues, we propose Base model Oriented Systematic Evaluation (BOSE), a method specifically designed to optimize the evaluation of base models. Specifically, BOSE introduces two key innovations: In-Context Light-instruction Prompt (ICLiP) for open-ended tasks and Blank-ppl for multi-choice tasks with candidate options, which transforms the standard perplexity (ppl) metric into a fill-in-the-blank format to mitigate early-stage evaluation fluctuations. Furthermore, we are the first to propose Kendall's rank correlation to quantitatively measure the evaluation stability and consistency. Experimental results demonstrate that BOSE significantly enhances both the stability of evaluations during pre-training and the consistency between base and instruct models, thereby providing more reliable guidance for the LLMs' training.

We introduce the Agent GPA (Goal-Plan-Action) framework: an evaluation paradigm based on an agent's operational loop of setting goals, devising plans, and executing actions. The framework includes five evaluation metrics: Goal Fulfillment, Logical Consistency, Execution Efficiency, Plan Quality, and Plan Adherence. Logical Consistency checks that an agent's actions are consistent with its prior actions. Execution Efficiency checks whether the agent executes in the most efficient way to achieve its goal. Plan Quality checks whether an agent's plans are aligned with its goals; Plan Adherence checks if an agent's actions are aligned with its plan; and Goal Fulfillment checks that agent's final outcomes match the stated goals. Our experimental results on two benchmark datasets - the public TRAIL/GAIA dataset and an internal dataset for a production-grade data agent - show that this framework (a) provides a systematic way to cover a broad range of agent failures, including all agent errors on the TRAIL/GAIA benchmark dataset; (b) supports LLM-judges that exhibit strong agreement with human annotation, covering 80% to over 95% errors; and (c) localizes errors with 86% agreement to enable targeted improvement of agent performance.

This research introduces the Judge's Verdict Benchmark, a novel two-step methodology to evaluate Large Language Models (LLMs) as judges for response accuracy evaluation tasks. We assess how well 54 LLMs can replicate human judgment when scoring responses from RAG (Retrieval-Augmented Generation) or Agentic pipelines against ground truth answers. Our methodology progresses from traditional correlation analysis to comprehensive Cohen's Kappa analysis that measures actual agreement patterns. The two-step approach includes: (1) a correlation test that filters judges with strong alignment, followed by (2) a human-likeness test using z-scores to identify two distinct judgment patterns: human-like judgment (|z| < 1) that mimics natural human variation, and super-consistent judgment (z > 1) that exceeds typical human-to-human agreement levels. This methodology reveals that 27 out of 54 tested LLMs achieve Tier 1 performance: 23 models exhibit human-like patterns that preserve the nuances of human judgment, while 4 models demonstrate super-consistent behavior, a pattern that could indicate either enhanced reliability or oversimplification of complex judgments. Testing 43 open-source models (1B-405B parameters) and 11 closed models (GPT, Gemini, Claude variants), we demonstrate that judge excellence is not solely dependent on model size but on specific training strategies. Our key contributions include: (1) establishing that correlation alone is insufficient for judge evaluation, (2) introducing a "Turing Test for judges" based on agreement patterns, and (3) providing a standardized benchmark for classifying LLM judges into distinct performance tiers for different evaluation needs.

When prompting SLMs for psychometric assessments, researchers assume the outputs reflect semantic reasoning. We evaluate this premise across 13 open-weights models (0.6B to 14B parameters) using a prompt variation framework that separates semantic signals from prompt artifacts. By systematically varying personas, instructions, items, and option symbols, we find that artifactual variance frequently overpowers the semantic signal. In these cases, models predominantly reflect prompt compliance rather than simulated psychological traits. While these findings limit SLM utility in psychometrics, our framework provides a diagnostic tool to identify destructive artifacts and isolate semantic understanding for future frontier-model research.

Any-to-any generative models aim to enable seamless interpretation and generation across multiple modalities within a unified framework, yet their ability to preserve relationships across modalities remains uncertain. Do unified models truly achieve cross-modal coherence, or is this coherence merely perceived? To explore this, we introduce ACON, a dataset of 1,000 images (500 newly contributed) paired with captions, editing instructions, and Q&A pairs to evaluate cross-modal transfers rigorously. Using three consistency criteria-cyclic consistency, forward equivariance, and conjugated equivariance-our experiments reveal that any-to-any models do not consistently demonstrate greater cross-modal consistency than specialized models in pointwise evaluations such as cyclic consistency. However, equivariance evaluations uncover weak but observable consistency through structured analyses of the intermediate latent space enabled by multiple editing operations. We release our code and data at https://github.com/JiwanChung/ACON.

Large Language Models (LLMs) are extensively used today across various sectors, including academia, research, business, and finance, for tasks such as text generation, summarization, and translation. Despite their widespread adoption, these models often produce incorrect and misleading information, exhibiting a tendency to hallucinate. This behavior can be attributed to several factors, with consistency and reasoning capabilities being significant contributors. LLMs frequently lack the ability to generate explanations and engage in coherent reasoning, leading to inaccurate responses. Moreover, they exhibit inconsistencies in their outputs. This paper aims to evaluate and compare the consistency and reasoning capabilities of both public and proprietary LLMs. The experiments utilize the Boolq dataset as the ground truth, comprising questions, answers, and corresponding explanations. Queries from the dataset are presented as prompts to the LLMs, and the generated responses are evaluated against the ground truth answers. Additionally, explanations are generated to assess the models' reasoning abilities. Consistency is evaluated by repeatedly presenting the same query to the models and observing for variations in their responses. For measuring reasoning capabilities, the generated explanations are compared to the ground truth explanations using metrics such as BERT, BLEU, and F-1 scores. The findings reveal that proprietary models generally outperform public models in terms of both consistency and reasoning capabilities. However, even when presented with basic general knowledge questions, none of the models achieved a score of 90\% in both consistency and reasoning. This study underscores the direct correlation between consistency and reasoning abilities in LLMs and highlights the inherent reasoning challenges present in current language models.

Evaluating large language models (LLMs) relies heavily on benchmark scores, yet aggregate metrics can obscure whether individual items serve their intended measurement purpose. BSDProbe provides an item-level framework for benchmark structural diagnosis that jointly characterizes correctness transitions and their supporting response evidence. Repeated responses along an empirically ordered model axis yield complementary correctness and normalized entropy trajectories, from which four-dimensional profiles of boundary position, boundary width, boundary-signal validity, and order consistency are constructed for benchmark-level aggregation. Experiments on six benchmarks identify varying proportions of potentially anomalous items. With Qwen3 as both the selection and reference axis, BSDProbe achieves a 17.9\% mean relative improvement in normalized discrimination ratio over the best baseline per benchmark and matches or exceeds all baselines in ranking agreement with the full benchmark.

In the field of psychology, traditional assessment methods, such as standardized scales, are frequently critiqued for their static nature, lack of personalization, and reduced participant engagement, while comprehensive counseling evaluations are often inaccessible. The complexity of quantifying psychological traits further limits these methods. Despite advances with large language models (LLMs), many still depend on single-round Question-and-Answer interactions. To bridge this gap, we introduce PsyDI, a personalized and progressively in-depth chatbot designed for psychological measurements, exemplified by its application in the Myers-Briggs Type Indicator (MBTI) framework. PsyDI leverages user-related multi-modal information and engages in customized, multi-turn interactions to provide personalized, easily accessible measurements, while ensuring precise MBTI type determination. To address the challenge of unquantifiable psychological traits, we introduce a novel training paradigm that involves learning the ranking of proxy variables associated with these traits, culminating in a robust score model for MBTI measurements. The score model enables PsyDI to conduct comprehensive and precise measurements through multi-turn interactions within a unified estimation context. Through various experiments, we validate the efficacy of both the score model and the PsyDI pipeline, demonstrating its potential to serve as a general framework for psychological measurements. Furthermore, the online deployment of PsyDI has garnered substantial user engagement, with over 3,000 visits, resulting in the collection of numerous multi-turn dialogues annotated with MBTI types, which facilitates further research.

This study investigates the consistency of feedback ratings generated by OpenAI's GPT-4, a state-of-the-art artificial intelligence language model, across multiple iterations, time spans and stylistic variations. The model rated responses to tasks within the Higher Education (HE) subject domain of macroeconomics in terms of their content and style. Statistical analysis was conducted in order to learn more about the interrater reliability, consistency of the ratings across iterations and the correlation between ratings in terms of content and style. The results revealed a high interrater reliability with ICC scores ranging between 0.94 and 0.99 for different timespans, suggesting that GPT-4 is capable of generating consistent ratings across repetitions with a clear prompt. Style and content ratings show a high correlation of 0.87. When applying a non-adequate style the average content ratings remained constant, while style ratings decreased, which indicates that the large language model (LLM) effectively distinguishes between these two criteria during evaluation. The prompt used in this study is furthermore presented and explained. Further research is necessary to assess the robustness and reliability of AI models in various use cases.

Large language models (LLMs) can explain their predictions through post-hoc or Chain-of-Thought (CoT) explanations. But an LLM could make up reasonably sounding explanations that are unfaithful to its underlying reasoning. Recent work has designed tests that aim to judge the faithfulness of post-hoc or CoT explanations. In this work we argue that these faithfulness tests do not measure faithfulness to the models' inner workings -- but rather their self-consistency at output level. Our contributions are three-fold: i) We clarify the status of faithfulness tests in view of model explainability, characterising them as self-consistency tests instead. This assessment we underline by ii) constructing a Comparative Consistency Bank for self-consistency tests that for the first time compares existing tests on a common suite of 11 open LLMs and 5 tasks -- including iii) our new self-consistency measure CC-SHAP. CC-SHAP is a fine-grained measure (not a test) of LLM self-consistency. It compares how a model's input contributes to the predicted answer and to generating the explanation. Our fine-grained CC-SHAP metric allows us iii) to compare LLM behaviour when making predictions and to analyse the effect of other consistency tests at a deeper level, which takes us one step further towards measuring faithfulness by bringing us closer to the internals of the model than strictly surface output-oriented tests. Our code is available at https://github.com/Heidelberg-NLP/CC-SHAP

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).

Simulating human clients in mental health counseling is crucial for training and evaluating counselors (both human or simulated) in a scalable manner. Nevertheless, past research on client simulation did not focus on complex conversation tasks such as mental health counseling. In these tasks, the challenge is to ensure that the client's actions (i.e., interactions with the counselor) are consistent with with its stipulated profiles and negative behavior settings. In this paper, we propose a novel framework that supports consistent client simulation for mental health counseling. Our framework tracks the mental state of a simulated client, controls its state transitions, and generates for each state behaviors consistent with the client's motivation, beliefs, preferred plan to change, and receptivity. By varying the client profile and receptivity, we demonstrate that consistent simulated clients for different counseling scenarios can be effectively created. Both our automatic and expert evaluations on the generated counseling sessions also show that our client simulation method achieves higher consistency than previous methods.

Stylistic personalization - making LLMs write in a specific individual's style, rather than merely adapting to task preferences - lacks evaluation grounded in authorship science. We show that grounding evaluation in authorship verification theory transforms what benchmarks can measure. Drawing on three measurement traditions - LUAR, a trained authorship verification model; an LLM-as-judge with decoupled trait matching; and classical function-word stylometrics - we evaluate four inference-time personalization methods across 50 authors and 1,000 generations. The theory-grounded metric, LUAR, provides what ad hoc alternatives cannot: calibrated baselines, with a human ceiling of 0.756 and a cross-author floor of 0.626, that give scores absolute meaning. All methods score below this floor, from 0.484 to 0.508, exposing an authorship gap invisible to uncalibrated metrics. The three metrics produce near-zero pairwise correlations, with absolute r less than 0.07, confirming that without theoretical grounding, metric choice determines conclusions: an LLM judge declares a clear winner while LUAR finds no meaningful differentiation. These findings demonstrate the theory-benchmark cycle in action: authorship theory exposes evaluation failures that ad hoc benchmarks miss.

Exam-style accuracy does not establish whether large language models (LLMs) reason well over clinical records. We define clinical reasoning as integrating and updating evidence across time and sources to form, revise and justify a patient's problem representation and a defensible plan. This structured narrative review maps three literatures: medical education assessment instruments, clinical LLM benchmarks published from 2023 onwards, and general-domain methods for evaluating long-form generation. We examine six dimensions: problem representation, temporal synthesis, differential and management reasoning, counterfactual reasoning, calibrated uncertainty, and reasoning faithfulness. Preprints are included and flagged. No single instrument covers all six dimensions. Problem representation and differential or management reasoning are reasonably covered, although reliability varies by instrument and setting. TIMER-Eval targets temporal synthesis, and ER-Reason assesses sequential diagnostic belief updating. Dedicated uncertainty and counterfactual evaluations are emerging, but their applicability to longitudinal free-text reasoning remains limited. Factual completeness is well theorised in general-domain evaluation, with early clinical evidence of important omissions. Faithfulness remains the weakest dimension, with one identified clinical causal-ablation study on multiple-choice questions. Existing tools should be combined through binary rubric items, separate completeness and correctness scores, case-specific importance weighting with non-compensable safety caps, temporal order-consistency checks, and chance-corrected reliability reporting. Further design work is needed for calibrated uncertainty, counterfactual reasoning and faithfulness over longitudinal free-text records. This review provides a design rationale, not a validated instrument.

Human values and their measurement are long-standing interdisciplinary inquiry. Recent advances in AI have sparked renewed interest in this area, with large language models (LLMs) emerging as both tools and subjects of value measurement. This work introduces Generative Psychometrics for Values (GPV), an LLM-based, data-driven value measurement paradigm, theoretically grounded in text-revealed selective perceptions. We begin by fine-tuning an LLM for accurate perception-level value measurement and verifying the capability of LLMs to parse texts into perceptions, forming the core of the GPV pipeline. Applying GPV to human-authored blogs, we demonstrate its stability, validity, and superiority over prior psychological tools. Then, extending GPV to LLM value measurement, we advance the current art with 1) a psychometric methodology that measures LLM values based on their scalable and free-form outputs, enabling context-specific measurement; 2) a comparative analysis of measurement paradigms, indicating response biases of prior methods; and 3) an attempt to bridge LLM values and their safety, revealing the predictive power of different value systems and the impacts of various values on LLM safety. Through interdisciplinary efforts, we aim to leverage AI for next-generation psychometrics and psychometrics for value-aligned AI.

State-of-the-art language models (LMs) are notoriously susceptible to generating hallucinated information. Such inaccurate outputs not only undermine the reliability of these models but also limit their use and raise serious concerns about misinformation and propaganda. In this work, we focus on hallucinated book and article references and present them as the "model organism" of language model hallucination research, due to their frequent and easy-to-discern nature. We posit that if a language model cites a particular reference in its output, then it should ideally possess sufficient information about its authors and content, among other relevant details. Using this basic insight, we illustrate that one can identify hallucinated references without ever consulting any external resources, by asking a set of direct or indirect queries to the language model about the references. These queries can be considered as "consistency checks." Our findings highlight that while LMs, including GPT-4, often produce inconsistent author lists for hallucinated references, they also often accurately recall the authors of real references. In this sense, the LM can be said to "know" when it is hallucinating references. Furthermore, these findings show how hallucinated references can be dissected to shed light on their nature. Replication code and results can be found at https://github.com/microsoft/hallucinated-references.

As LLM benchmarks saturate, the evaluation community has pursued two strategies to increase difficulty: escalating knowledge demands (GPQA, HLE) or removing knowledge entirely in favor of abstract reasoning (ARC-AGI). The first conflates memorization with capability; the second divorces reasoning from the practical contexts in which it matters. We take a different approach. The Grounded Integration Measure (GIM) is a benchmark of 820 original problems (615 public, 205 private) where difficulty comes from integration; individual problems require coordinating multiple cognitive operations (constraint satisfaction, state tracking, epistemic vigilance, audience calibration) over broadly accessible knowledge, so that reasoning stays grounded in realistic tasks without being gated on specialized expertise. Each problem is an original expert-authored composition, majority with rubric-decomposed scoring (median 6 independently judged criteria). A balanced public--private split provides built-in contamination diagnostic. We calibrate a continuous response 2-parameter logistic (2PL) IRT model over >200k prompt-response pairs across 28 models, producing robust ability estimates that correctly order test-configurations even when raw accuracy is distorted by errors or missing data, addressing a common challenge in benchmark reporting. Using this framework, we present a comprehensive leaderboard spanning 22 models and 47 test-configurations (unique model, thinking-level pairs), and conduct what is to our knowledge the most extensive published study of how test-time compute trades off against model capability on a fixed benchmark: 11 models swept across 35 test-configurations. We observe that within-family configuration choices, such as thinking budget and quantization, matter as much as model selection. We release the evaluation framework, calibrated IRT parameters, and all public problems.

Tracking depression from multi-session counseling dialogues requires estimating both current symptom severity and how it changes across sessions. Yet progress on this task is constrained by the scarcity of longitudinal counseling data with standardized session-level depression labels. Existing resources typically provide either multi-session conversations without depression labels or labeled interviews in a single session. Building such a benchmark poses three challenges: maintaining longitudinal consistency and diversity, grounding symptom progression in empirical patterns, and expressing controlled depression states naturally without exposing target labels. To address these challenges, we introduce LongCounsel-8, a benchmark suite of three independently generated datasets totaling 7,749 five-session counseling trajectories, grounded in real-world client profiles, depression trajectories, symptom compositions, and counseling patterns. We combine profile-grounded simulation, empirically informed state construction, and indirect behavioral realization to address these challenges. Across the benchmark, simulated self-reports closely recover the controlled states, supporting label fidelity. Experiments on existing depression tracking methods reveal three key findings: (1) lower single-session score error does not guarantee accurate identification of trend, i.e., improvement or worsening; (2) existing methods are consistently less reliable on worsening trajectories; and (3) additional session history may reduce the accuracy of trend prediction. Together, these findings establish LongCounsel-8 as a foundation for advancing depression assessment from static, single-session prediction toward reliable longitudinal tracking of mental-health change.

In Natural Language Processing (NLP), the Elo rating system, originally designed for ranking players in dynamic games such as chess, is increasingly being used to evaluate Large Language Models (LLMs) through "A vs B" paired comparisons. However, while popular, the system's suitability for assessing entities with constant skill levels, such as LLMs, remains relatively unexplored. We study two fundamental axioms that evaluation methods should adhere to: reliability and transitivity. We conduct extensive evaluation of Elo behaviour, illustrating that individual Elo computations exhibit volatility and delving into the impact of varying the Elo rating system's hyperparameters. We show that these axioms are not always satisfied raising questions about the reliability of current comparative evaluations of LLMs. If the current use of Elo scores is intended to substitute the costly head-to-head comparison of LLMs, it is crucial to ensure the ranking is as robust as possible. Guided by the axioms, our findings offer concrete guidelines for enhancing the reliability of LLM evaluation methods, suggesting a need for reassessment of existing comparative approaches.

LLM-as-a-judge has become the dominant paradigm for grading model outputs at scale, yet the same model assigns systematically different scores when its output format changes (e.g., a 1-5 rating vs. a True/False label). Existing diagnoses of these format-induced inconsistencies stop at the input-output level. Using Position-aware Edge Attribution Patching (PEAP), we causally investigate the internal mechanism in Gemma-3, Qwen2.5, and Llama-3. We find that judgments across structured understanding and open-ended preference tasks share a sparse, generalized Latent Evaluator sub-graph in the mid-to-late multi-layer perceptrons (MLPs); zero-ablating it collapses judgment while preserving world knowledge in architecturally modular models. By structurally decoupling abstract judging from output formatting, we provide a mechanistic account of format-induced inconsistency on the open-weight models we study: a continuous judgment signal computed in the shared trunk is mapped through fragile, format-specific terminal branches, enabling format-independent preference to be isolated downstream of the requested output format. Our findings imply that benchmark-level reliability comparisons across formats are partially measuring formatter geometry rather than evaluation quality.

Current preference learning methods achieve high accuracy on standard benchmarks but exhibit significant performance degradation when objective quality signals are removed. We introduce WritingPreferenceBench, a dataset of 1,800 human-annotated preference pairs (1,200 English, 600 Chinese) across 8 creative writing genres, where responses are matched for objective correctness, factual accuracy, and length. On this benchmark, sequence-based reward models--the standard architecture for RLHF--achieve only 52.7% mean accuracy, while zero-shot language model judges perform at 53.9%. In contrast, generative reward models that produce explicit reasoning chains achieve 81.8% accuracy. We observe high within-model variance across genres: individual models range from 18.2% to 81.8% accuracy across different writing categories, with standard deviations averaging 10.1%. This variance persists regardless of model scale, with 27B parameter models showing no consistent improvement over 8B variants. Our results suggest that current RLHF methods primarily learn to detect objective errors rather than capture subjective quality preferences (e.g., creativity, stylistic flair, and emotional resonance), and that successful preference modeling may require intermediate reasoning representations rather than direct classification.

Sycophancy in large language models risks amplifying unsupported claims or beliefs across the knowledge ecosystem as these systems increasingly contribute to research synthesis, evaluation, and decision-making. A reliable model should revise its judgement when a user identifies a genuine flaw, but should resist pushback based on fallacies, emotion, or irrelevant appeals to authority. While sycophancy is usually measured on questions with a verifiable answer, the evaluative tasks in which it is most difficult to detect have no such ground truth. We introduce Big-Big-SycoBench, which measures sycophancy through argument discernment: the ability to revise an evaluation after valid criticism while resisting invalid pressure. The benchmark consists of 450 text artefacts with 5,400 arguments for raising or lowering their quality scores, each reviewed by one or more of twelve human annotators. An LLM scores each artefact twenty times to establish its baseline score and uncertainty. The model is then given a valid or invalid argument and asked to update its score. We only count a revision if it exceeds the model's own baseline variability. We use the resulting score delta to distinguish responsiveness to valid arguments from compliance with invalid ones. Across ten models, responsiveness is 88-95\% for seven, 76-77\% for Kimi K3 and GLM-5.3, and 32\% for Opus 5.5. Compliance ranges from 0.2\% for GPT-6 Astra to 95\% for GPT-4o. Big-Big-SycoBench is intended to guide the development of models whose judgements can be corrected by sound arguments, but not displaced by unsupported pressure.

The rapid release of both language models and benchmarks makes it increasingly costly to evaluate every model on every dataset. In practice, models are often evaluated on different samples, making scores difficult to compare across studies. To address this, we propose a framework based on multidimensional Item Response Theory (IRT) that uses anchor items to calibrate new benchmarks to the evaluation suite while holding previously calibrated item parameters fixed. Our approach supports a realistic evaluation setting in which datasets are introduced over time and models are evaluated only on the datasets available at the time of evaluation, while a fixed anchor set for each dataset is used so that results from different evaluation periods can be compared directly. In large-scale experiments on more than 400 models, our framework predicts full-evaluation performance within 2-3 percentage points using only 100 anchor questions per dataset, with Spearman ρgeq 0.9 for ranking preservation, showing that it is possible to extend benchmark suites over time while preserving score comparability, at a constant evaluation cost per new dataset. Code available at https://github.com/eliyahabba/growing-pains

Ensuring persona fidelity in large language models (LLMs) is essential for maintaining coherent and engaging human-AI interactions. However, LLMs often exhibit Out-of-Character (OOC) behavior, where generated responses deviate from an assigned persona, leading to inconsistencies that affect model reliability. Existing evaluation methods typically assign single scores to entire responses, struggling to capture subtle persona misalignment, particularly in long-form text generation. To address this limitation, we propose an atomic-level evaluation framework that quantifies persona fidelity at a finer granularity. Our three key metrics measure the degree of persona alignment and consistency within and across generations. Our approach enables a more precise and realistic assessment of persona fidelity by identifying subtle deviations that real users would encounter. Through our experiments, we demonstrate that our framework effectively detects persona inconsistencies that prior methods overlook. By analyzing persona fidelity across diverse tasks and personality types, we reveal how task structure and persona desirability influence model adaptability, highlighting challenges in maintaining consistent persona expression.

AI psychometrics evaluates AI systems in roles that traditionally require emotional judgment and ethical consideration. Prior work often reuses human trait inventories (Big Five, \hexaco) or ad hoc personas, limiting behavioral realism and domain relevance. We propose a framework that (1) uses situational judgment tests (SJTs) from realistic scenarios to probe domain-specific competencies; (2) integrates industrial-organizational and personality psychology to design sophisticated personas which include behavioral and psychological descriptors, life history, and social and emotional functions; and (3) employs structured generation with population demographic priors and memoir inspired narratives, encoded with Pydantic schemas. In a law enforcement assistant case study, we construct a rich dataset of personas drawn across 8 persona archetypes and SJTs across 11 attributes, and analyze behaviors across subpopulation and scenario slices. The dataset spans 8,500 personas, 4,000 SJTs, and 300,000 responses. We will release the dataset and all code to the public.

As large language models (LLM) evolve in their capabilities, various recent studies have tried to quantify their behavior using psychological tools created to study human behavior. One such example is the measurement of "personality" of LLMs using self-assessment personality tests developed to measure human personality. Yet almost none of these works verify the applicability of these tests on LLMs. In this paper, we analyze the reliability of LLM personality scores obtained from self-assessment personality tests using two simple experiments. We first introduce the property of prompt sensitivity, where three semantically equivalent prompts representing three intuitive ways of administering self-assessment tests on LLMs are used to measure the personality of the same LLM. We find that all three prompts lead to very different personality scores, a difference that is statistically significant for all traits in a large majority of scenarios. We then introduce the property of option-order symmetry for personality measurement of LLMs. Since most of the self-assessment tests exist in the form of multiple choice question (MCQ) questions, we argue that the scores should also be robust to not just the prompt template but also the order in which the options are presented. This test unsurprisingly reveals that the self-assessment test scores are not robust to the order of the options. These simple tests, done on ChatGPT and three Llama2 models of different sizes, show that self-assessment personality tests created for humans are unreliable measures of personality in LLMs.

LLM-as-a-Judge has been widely adopted as an evaluation method and served as supervised rewards in model training. However, existing benchmarks for LLM-as-a-Judge are mainly relying on human-annotated ground truth, which introduces human bias that undermines the assessment of reliability and imposes scalability constraints. To overcome these limitations, we introduce Sage, a novel evaluation suite that assesses the quality of LLM judges without necessitating any human annotation. Inspired by axioms of rational choice theory, Sage introduces two new lenses for measuring LLM-as-a-Judge: local self-consistency (pair-wise preference stability) and global logical consistency (transitivity across a full set of preferences). We curate a dataset of 650 questions by combining structured benchmark problems with real-world user queries. Our experiments demonstrate both the stability of our metrics and their high correlation with supervised benchmarks like LLMBar and RewardBench2, confirming Sage's reliability as an evaluation suite for the robustness and accuracy of LLM-as-a-Judge. Based on Sage, we reveal that current state-of-the-art LLMs exhibit significant reliability problems when acting as judges in both scoring and pairwise settings; even the top-performing models, Gemini-2.5-Pro and GPT-5, fail to maintain consistent preferences in nearly a quarter of difficult cases. We attribute this to a new phenomenon called situational preference, which explains why explicit rubrics or criteria can help the model judge consistently across answer pairs. Our further analysis shows that finetuned LLM-as-a-Judge is a feasible method to boost performance, and the panel-based judge as well as deep reasoning can enhance the judging consistency. We also find substantial inconsistency in human judgments, which indicates that human annotation may not be a reliable gold standard.

Large language models are increasingly deployed as automated judges for evaluating other models, yet the stability of their verdicts under semantically equivalent prompt paraphrases remains unmeasured. We introduce JudgeSense, a framework and benchmark for quantifying this property via the Judge Sensitivity Score (JSS), defined as the fraction of paraphrase pairs on which a judge returns an identical decision. Evaluating nine judge models on 494 validated paraphrase pairs, we find that coherence is the only task where judges meaningfully differ, with JSS ranging from 0.389 to 0.992. On factuality, all judges cluster near JSS about 0.63, driven by a polarity-inverted prompt artifact; after correction, factuality JSS rises to about 0.9. Pairwise tasks (preference and relevance) exhibit degenerate always-A behavior in 8 of 9 judges, indicating strong position bias. Model scale does not predict consistency. We release code, decision logs, and a validated paraphrase dataset to support standardized JSS reporting.

Personality traits have long been studied as predictors of human behavior. Recent advances in Large Language Models (LLMs) suggest similar patterns may emerge in artificial systems, with advanced LLMs displaying consistent behavioral tendencies resembling human traits like agreeableness and self-regulation. Understanding these patterns is crucial, yet prior work primarily relied on simplified self-reports and heuristic prompting, with little behavioral validation. In this study, we systematically characterize LLM personality across three dimensions: (1) the dynamic emergence and evolution of trait profiles throughout training stages; (2) the predictive validity of self-reported traits in behavioral tasks; and (3) the impact of targeted interventions, such as persona injection, on both self-reports and behavior. Our findings reveal that instructional alignment (e.g., RLHF, instruction tuning) significantly stabilizes trait expression and strengthens trait correlations in ways that mirror human data. However, these self-reported traits do not reliably predict behavior, and observed associations often diverge from human patterns. While persona injection successfully steers self-reports in the intended direction, it exerts little or inconsistent effect on actual behavior. By distinguishing surface-level trait expression from behavioral consistency, our findings challenge assumptions about LLM personality and underscore the need for deeper evaluation in alignment and interpretability.

Human-like personality traits have recently been discovered in large language models, raising the hypothesis that their (known and as yet undiscovered) biases conform with human latent psychological constructs. While large conversational models may be tricked into answering psychometric questionnaires, the latent psychological constructs of thousands of simpler transformers, trained for other tasks, cannot be assessed because appropriate psychometric methods are currently lacking. Here, we show how standard psychological questionnaires can be reformulated into natural language inference prompts, and we provide a code library to support the psychometric assessment of arbitrary models. We demonstrate, using a sample of 88 publicly available models, the existence of human-like mental health-related constructs (including anxiety, depression, and Sense of Coherence) which conform with standard theories in human psychology and show similar correlations and mitigation strategies. The ability to interpret and rectify the performance of language models by using psychological tools can boost the development of more explainable, controllable, and trustworthy models.

Language model (LM) benchmarking faces several challenges: comprehensive evaluations are costly, benchmarks often fail to measure the intended capabilities, and evaluation quality can degrade due to labeling errors and benchmark saturation. Although various strategies have been proposed to mitigate these issues, they tend to address individual aspects in isolation, neglecting broader questions about overall evaluation quality. Here, we introduce Fluid Benchmarking, a new evaluation approach that advances LM benchmarking across multiple dimensions. Inspired by psychometrics, Fluid Benchmarking is based on the insight that the relative value of benchmark items depends on an LM's capability level, suggesting that evaluation should adapt to each LM. Methodologically, Fluid Benchmarking estimates an item response model based on existing LM evaluation results and uses the inferred quantities to select evaluation items dynamically, similar to computerized adaptive testing in education. In our experiments, we compare Fluid Benchmarking against the common practice of random item sampling as well as more sophisticated baselines, including alternative methods grounded in item response theory. We examine four dimensions -- efficiency, validity, variance, and saturation -- and find that Fluid Benchmarking achieves superior performance in all of them (e.g., higher validity and less variance on MMLU with fifty times fewer items). Our analysis shows that the two components of Fluid Benchmarking have distinct effects: item response theory, used to map performance into a latent ability space, increases validity, while dynamic item selection reduces variance. Overall, our results suggest that LM benchmarking can be substantially improved by moving beyond static evaluation.

Medical large language models (LLMs) research often makes bold claims, from encoding clinical knowledge to reasoning like a physician. These claims are usually backed by evaluation on competitive benchmarks; a tradition inherited from mainstream machine learning. But how do we separate real progress from a leaderboard flex? Medical LLM benchmarks, much like those in other fields, are arbitrarily constructed using medical licensing exam questions. For these benchmarks to truly measure progress, they must accurately capture the real-world tasks they aim to represent. In this position paper, we argue that medical LLM benchmarks should (and indeed can) be empirically evaluated for their construct validity. In the psychological testing literature, "construct validity" refers to the ability of a test to measure an underlying "construct", that is the actual conceptual target of evaluation. By drawing an analogy between LLM benchmarks and psychological tests, we explain how frameworks from this field can provide empirical foundations for validating benchmarks. To put these ideas into practice, we use real-world clinical data in proof-of-concept experiments to evaluate popular medical LLM benchmarks and report significant gaps in their construct validity. Finally, we outline a vision for a new ecosystem of medical LLM evaluation centered around the creation of valid benchmarks.

Factual consistency is an essential quality of text summarization models in practical settings. Existing work in evaluating this dimension can be broadly categorized into two lines of research, entailment-based and question answering (QA)-based metrics, and different experimental setups often lead to contrasting conclusions as to which paradigm performs the best. In this work, we conduct an extensive comparison of entailment and QA-based metrics, demonstrating that carefully choosing the components of a QA-based metric, especially question generation and answerability classification, is critical to performance. Building on those insights, we propose an optimized metric, which we call QAFactEval, that leads to a 14% average improvement over previous QA-based metrics on the SummaC factual consistency benchmark, and also outperforms the best-performing entailment-based metric. Moreover, we find that QA-based and entailment-based metrics can offer complementary signals and be combined into a single metric for a further performance boost.

Multimodal Large Language Models (MLLMs) are increasingly deployed in human-facing roles where personality perception is critical, yet existing benchmarks evaluate this capability solely on numerical Big Five score prediction, leaving open whether models truly perceive personality through behavioral understanding or merely prejudge through superficial pattern matching. We address this gap with three contributions. (i) A new task: we formalize Grounded Personality Reasoning (GPR), which requires MLLMs to anchor each Big Five rating in observable evidence through a chain of rating, reasoning, and grounding. (ii) A new dataset: we release MM-OCEAN (1,104 videos, 5,320 MCQs), produced by a multi-agent pipeline with human verification, with timestamped behavioral observations, evidence-grounded trait analyses, and seven categories of cue-grounding MCQs. (iii) Benchmark and analysis: we design a three-tier evaluation (rating, reasoning, grounding) plus four sample-level failure-mode metrics: Prejudice Rate (PR), Confabulation Rate (CR), Integration-failure Rate (IR), and Holistic-grounding Rate (HR), and benchmark 27 MLLMs (13 closed, 14 open). The analysis uncovers a striking Prejudice Gap: across the field, 51% of correct ratings are not grounded in retrieved cues, and the Holistic-Grounding Rate spans only 0-33.5%. These findings expose a disconnect between getting the right score and reasoning for the right reason, charting a roadmap for grounded social cognition in MLLMs.

Large Language Models (LLMs) have demonstrated remarkable capabilities, revolutionizing the integration of AI in daily life applications. However, they are prone to hallucinations, generating claims that contradict established facts, deviating from prompts, and producing inconsistent responses when the same prompt is presented multiple times. Addressing these issues is challenging due to the lack of comprehensive and easily assessable benchmark datasets. Most existing datasets are small and rely on multiple-choice questions, which are inadequate for evaluating the generative prowess of LLMs. To measure hallucination in LLMs, this paper introduces a comprehensive benchmark dataset comprising over 75,000 prompts across eight domains. These prompts are designed to elicit definitive, concise, and informative answers. The dataset is divided into two segments: one publicly available for testing and assessing LLM performance and a hidden segment for benchmarking various LLMs. In our experiments, we tested six LLMs-GPT-3.5, LLama 2, LLama 3, Gemini, Mixtral, and Zephyr-revealing that overall factual hallucination ranges from 59% to 82% on the public dataset and 57% to 76% in the hidden benchmark. Prompt misalignment hallucination ranges from 6% to 95% in the public dataset and 17% to 94% in the hidden counterpart. Average consistency ranges from 21% to 61% and 22% to 63%, respectively. Domain-wise analysis shows that LLM performance significantly deteriorates when asked for specific numeric information while performing moderately with person, location, and date queries. Our dataset demonstrates its efficacy and serves as a comprehensive benchmark for LLM performance evaluation. Our dataset and LLMs responses are available at https://github.com/ashikiut/DefAn{https://github.com/ashikiut/DefAn}.

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.

Evaluating the physical consistency of generated videos remains a fundamental challenge. Existing approaches rely on off-the-shelf vision-language models, which can often be myopic to physical dynamics, or fine-tuned evaluators trained on human annotations, which overfit to dataset-specific cues and fail to generalize. A key challenge is that existing supervision sources provide either relative ordering or absolute scores, but not both reliably and consistently across varied settings. To this end, we introduce PhyProbe, an evaluator that extracts features from a frozen pretrained spatio-temporal encoder and maps them to a scalar physical consistency violation score via a lightweight scoring head. PhyProbe is trained through a unified objective combining pairwise ranking, regression on noisy scalar annotations, and anchor-based calibration over a curated set of heterogeneous supervision sources. Experiments show that PhyProbe outperforms prior methods on most pairwise benchmarks spanning real-generated and generated-generated pairs under varying correspondence, with the largest gains in no-correspondence and generated-generated settings where existing fine-tuned evaluators degrade sharply. PhyProbe achieves strong correlation with human judgments, with close agreement between rank-based and linear metrics, indicating that scores are both well ordered and anchored to a stable [0, 1] scale. Further, despite being trained on supervision indicative of physical consistency, without explicit general-preference labels, PhyProbe also performs competitively on human preference benchmarks: consistent with the observation that physics violations are entangled with broader quality degradations.

Diffusion models are relatively easy to train but require many steps to generate samples. Consistency models are far more difficult to train, but generate samples in a single step. In this paper we propose Multistep Consistency Models: A unification between Consistency Models (Song et al., 2023) and TRACT (Berthelot et al., 2023) that can interpolate between a consistency model and a diffusion model: a trade-off between sampling speed and sampling quality. Specifically, a 1-step consistency model is a conventional consistency model whereas we show that a infty-step consistency model is a diffusion model. Multistep Consistency Models work really well in practice. By increasing the sample budget from a single step to 2-8 steps, we can train models more easily that generate higher quality samples, while retaining much of the sampling speed benefits. Notable results are 1.4 FID on Imagenet 64 in 8 step and 2.1 FID on Imagenet128 in 8 steps with consistency distillation. We also show that our method scales to a text-to-image diffusion model, generating samples that are very close to the quality of the original model.

Large language models exhibit recurring behavioural patterns -- personas -- that shape generalisation and safety, but we lack reliable tools for decomposing, measuring, and controlling them. Our central insight is to treat personas as positions in a space of behavioural traits, using the OCEAN framework to describe model personas in terms of Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. We train low-rank adapters to amplify or suppress individual traits, and evaluate their effects using an LLM-judge calibrated against a human-validated panel, trait-specific multiple-choice benchmarks, and standard capability evaluations. Across six models from three families (4B-32B), we find that each adapter moves its target trait largely monotonically with scale, combines approximately additively with other adapters to construct mixed personas, and preserves performance on capability benchmarks at moderate scales. We further show that the induced trait axes affect safety-relevant behaviour in downstream evaluations: for example, moving along neuroticism and agreeableness axes affects frustration and sycophancy respectively. We also introduce an unsupervised psychometric pipeline that recovers four interpretable behavioural factors (tone, initiative, didacticism, epistemic caution) from model rollouts. Persona control can then be considered in terms of learning, scaling, and composing traits in weight space, providing a bridge between personality measurement, model editing, and safety.

Self-consistency (SC), a widely used decoding strategy for chain-of-thought reasoning, shows significant gains across various multi-step reasoning tasks but comes with a high cost due to multiple sampling with the preset size. Its variants, Adaptive self-consistency (ASC) and Early-stopping self-consistency (ESC), dynamically adjust the number of samples based on the posterior distribution of a set of pre-samples, reducing the cost of SC with minimal impact on performance. Both methods, however, do not exploit the prior information about question difficulty. It often results in unnecessary repeated sampling for easy questions that could be accurately answered with just one attempt, wasting resources. To tackle this problem, we propose Difficulty-Adaptive Self-Consistency (DSC), which leverages the difficulty information from both prior and posterior perspectives to adaptively allocate inference resources, further reducing the cost of SC. To demonstrate the effectiveness of DSC, we conduct extensive experiments on three popular categories of reasoning tasks: arithmetic, commonsense and symbolic reasoning on six benchmarks. The empirical results show that DSC consistently surpasses the strong baseline ASC and ESC in terms of costs by a significant margin, while attaining comparable performances.

A popular approach for improving the correctness of output from large language models (LLMs) is Self-Consistency - poll the LLM multiple times and output the most frequent solution. Existing Self-Consistency techniques always draw a constant number of samples per question, where a better approach will be to non-uniformly distribute the available budget based on the amount of agreement in the samples drawn so far. In response, we introduce Adaptive-Consistency, a cost-efficient, model-agnostic technique that dynamically adjusts the number of samples per question using a lightweight stopping criterion. Our experiments over 13 datasets and two LLMs demonstrate that Adaptive-Consistency reduces sample budget by up to 6.0 times with an average accuracy drop of less than 0.1%.

Modern AI systems depend on annotated data, but annotation is rarely treated as the act of measurement that it is. Instead, annotation quality is commonly reduced to agreement: if multiple annotators assign the same annotation to a data instance, the annotations are taken to be high-quality. Yet agreement does not establish whether annotations validly capture the underlying concept they are meant to represent. In this paper, we argue that data annotation should be understood as a measurement problem. Like other forms of measurement, annotation requires defining a concept, operationalizing it through an instrument, applying that instrument, and evaluating the reliability and validity of the resulting measurements. Drawing on a literature review of annotation quality research (N=132) and semi-structured interviews with annotation team members (N=10), we develop a framework for diagnosing and correcting annotation issues. First, we map key decision points across annotation processes - including task design, annotator management, quality assessment, quality improvement, and adjudication - that shape annotation outcomes. Second, we identify five distinct sources of annotation issues: error, ambiguity, impossibility, subjectivity, and annotator identity. Annotation problems that appear similar at the level of outcomes often require different process-level interventions based on their sources. Finally, we translate measurement theory into practical guidance for annotation teams, showing how assessments of reliability and validity can move beyond agreement alone. By reframing annotation as measurement, we offer a conceptual foundation for improving the quality of annotated data used in AI research and practice.

Indic quality estimation (QE) and automatic post-editing (APE) data is spread across separate releases, so no single resource supports training and evaluation across tasks and language pairs on one footing. We consolidate the WMT 2020--2024 shared-task lineage with an extended English--Malayalam resource into \indicqe: 126{,}754 instances over nine directional pairs, with up to four label types aligned on the same segment, a direct assessment, a human post-edit, word-level OK/BAD tags and an error explanation, and a test set stratified over four difficulty axes. On it, we benchmark six prompted LLMs and three COMET metrics on segment-level QE, and three systems on APE. Two of the axes are defined partly on the direct assessment and select a compressed slice of it, so each axis is compared against a control drawn from the same language pair with the same score distribution. Only one survives that control: segments whose holistic and token-level quality signals conflict are ranked worse than equally-scored segments of the same language, for all nine systems and all seven pairs that carry the axis. Annotator disagreement, which looks second-hardest without the control, has no effect with it. Few-shot prompting costs every model leq 3.4B both correlation and output-format compliance. Within-language accuracy does not make scores comparable across pairs: of the three trained metrics, the one with the best within-language correlation loses most when the pairs are pooled. The benchmark and code will be released.

The evaluation of LLMs has so far focused primarily on how well they can perform different tasks such as reasoning, question-answering, paraphrasing, or translating. For most of these tasks, performance can be measured with objective metrics, such as the number of correct answers. However, other language features are not easily quantified. For example, arousal, concreteness, or gender associated with a given word, as well as the extent to which we experience words with senses and relate them to a specific sense. Those features have been studied for many years by psycholinguistics, conducting large-scale experiments with humans to produce ratings for thousands of words. This opens an opportunity to evaluate how well LLMs align with human ratings on these word features, taking advantage of existing studies that cover many different language features in a large number of words. In this paper, we evaluate the alignment of a representative group of LLMs with human ratings on two psycholinguistic datasets: the Glasgow and Lancaster norms. These datasets cover thirteen features over thousands of words. The results show that alignment is black{generally} better in the Glasgow norms evaluated (arousal, valence, dominance, concreteness, imageability, familiarity, and gender) than on the Lancaster norms evaluated (introceptive, gustatory, olfactory, haptic, auditory, and visual). This suggests a potential limitation of current LLMs in aligning with human sensory associations for words, which may be due to their lack of embodied cognition present in humans and illustrates the usefulness of evaluating LLMs with psycholinguistic datasets.

Common evaluation paradigms for language models focus on scoring single responses through accuracy metrics or proper scoring rules, failing to capture the full richness of a model's belief state. Recent work illustrates that language models hallucinate in-part because they are optimised to be good test-takers under binary scoring schemes that reward any answer over abstention. While this insight naturally leads to penalty-based approaches, they ignore crucial distinctions in how models distribute uncertainty, for example between hedging toward incorrect answers versus hedging toward "I don't know" responses. A novel evaluation metric, the Distributional Correctness Score (DCS), is introduced to solve this problem, i.e., of not considering a model's entire probability distribution over answer choices. DCS naturally distinguishes between harmful overconfidence in wrong answers and uncertainty expressed through abstention, providing scores in an interpretable default range. Through theoretical analysis and illustrative examples, DCS is demonstrated to offer a more nuanced and aligned evaluation paradigm that incentivises models to express genuine uncertainty rather than guessing. Adapting 12 existing evaluation benchmarks to DCS's variants and measuring performance on six language models reveals that for half of the tested benchmarks scores are negative across all tested models, indicating significant tendencies towards hallucination.

Large language model (LLM) agents may perform well on isolated tasks yet drift into inconsistency over extended interaction. We evaluate temporal consistency in a controlled 20-step multi-agent setting inspired by delayed-gratification studies. At each step, an agent chooses between continuing to delay a reward or claiming it immediately (terminating the episode). Across a full-factorial manipulation of social visibility (private vs public), persona stressors, and deliberation policy, we run 84,540 trajectories spanning 8 model families. Treating the first reward-claim as a time-to-event outcome, we estimate Kaplan-Meier survival curves and fit discrete-time hazard regression to quantify how experimental factors shift failure risk over time. Then, to analyze rationales and language patterns associated with failure, we build a seven-category taxonomy from 13,780 deliberation traces from agents who choose to terminate the episode, using an LLM-assisted labeling paired with human audit (κ=0.83). Rationale profiles change systematically with time and context: early failures are more impulse-driven, later failures more fatigue- and cost-benefit-framed, while public settings increase norm-oriented justifications. We also find a deliberation-inconsistency association: among failures, longer deliberation correlates with higher rates of intra-rationale contradiction (simultaneous pro-delay and pro-claim statements), challenging the assumption that more reasoning text implies greater consistency. Together, the survival and rationale analyses reveal distinct temporal reliability regimes and model-specific "failure fingerprints", offering an evaluation lens for diagnosing inconsistency in multi-turn agent behavior.

We introduce a cross-domain behavioural assay of monitoring-control coupling in LLMs, grounded in the Nelson and Narens (1990) metacognitive framework and applying human psychometric methodology to LLM evaluation. The battery comprises 524 items across six cognitive domains (learning, metacognitive calibration, social cognition, attention, executive function, prospective regulation), each grounded in an established experimental paradigm. Tasks T1-T5 were pre-registered on OSF prior to data collection; T6 was added as an exploratory extension. After every forced-choice response, dual probes adapted from Koriat and Goldsmith (1996) ask the model to KEEP or WITHDRAW its answer and to BET or decline. The critical metric is the withdraw delta: the difference in withdrawal rate between incorrect and correct items. Applied to 20 frontier LLMs (10,480 evaluations), the battery discriminates three profiles consistent with the Nelson-Narens architecture: blanket confidence, blanket withdrawal, and selective sensitivity. Accuracy rank and metacognitive sensitivity rank are largely inverted. Retrospective monitoring and prospective regulation appear dissociable (r = .17, 95% CI wide given n=20; exemplar-based evidence is the primary support). Scaling on metacognitive calibration is architecture-dependent: monotonically decreasing (Qwen), monotonically increasing (GPT-5.4), or flat (Gemma). Behavioural findings converge structurally with an independent Type-2 SDT approach, providing preliminary cross-method construct validity. All items, data, and code: https://github.com/synthiumjp/metacognitive-monitoring-battery.

Language-model agents are increasingly deployed under standing instructions: a system prompt, a policy file, or a skills document is placed in context, and the agent is trusted to let that document govern every action that follows. Existing benchmarks rarely test this deployment pattern directly; they measure whether an agent can complete a task, not whether a long, binding policy document constrains its behavior over an extended tool-use horizon. We present HANDBOOK_md, a benchmark of 65 agentic tasks modeled on how employees follow company handbooks. Each task places an agent in a self-contained company environment (a file workspace with mock email, chat, calendar, issue-tracking, and commerce services exposed over the Model Context Protocol) and instructs it to carry out routine professional work governed by an expert-written standard operating procedure of 20-124 pages. Tasks span five domains (finance, medical billing, insurance, logistics, and HR) and 10 fictional companies. To resist memorization, every task modifies one of 10 base handbooks, altering the specific rules and thresholds on which grading depends, so no two tasks share the same set of policies. Grading is fully deterministic: each task carries a rubric of programmatic criteria (824 in total) that check both that required actions occurred and that prohibited actions did not. Under strict grading, where a trial passes only if every criterion is satisfied, the strongest evaluated model passes 36.2% of trials, and most frontier models remain below 25%. Failures follow consistent patterns: agents let a plausible but unauthorized in-environment request override the standing policy, perform a required check and then act against its result, lose rule details over long horizons, and report compliance they did not achieve. We release the tasks, environments, and evaluation harness.

Recent work has shown that models trained to the same objective, and which achieve similar measures of accuracy on consistent test data, may nonetheless behave very differently on individual predictions. This inconsistency is undesirable in high-stakes contexts, such as medical diagnosis and finance. We show that this inconsistent behavior extends beyond predictions to feature attributions, which may likewise have negative implications for the intelligibility of a model, and one's ability to find recourse for subjects. We then introduce selective ensembles to mitigate such inconsistencies by applying hypothesis testing to the predictions of a set of models trained using randomly-selected starting conditions; importantly, selective ensembles can abstain in cases where a consistent outcome cannot be achieved up to a specified confidence level. We prove that that prediction disagreement between selective ensembles is bounded, and empirically demonstrate that selective ensembles achieve consistent predictions and feature attributions while maintaining low abstention rates. On several benchmark datasets, selective ensembles reach zero inconsistently predicted points, with abstention rates as low 1.5%.

Patients often face difficulties in understanding their hospitalizations, while healthcare workers have limited resources to provide explanations. In this work, we investigate the potential of large language models to generate patient summaries based on doctors' notes and study the effect of training data on the faithfulness and quality of the generated summaries. To this end, we release (i) a rigorous labeling protocol for errors in medical texts and (ii) a publicly available dataset of annotated hallucinations in 100 doctor-written and 100 generated summaries. We show that fine-tuning on hallucination-free data effectively reduces hallucinations from 2.60 to 1.55 per summary for Llama 2, while preserving relevant information. We observe a similar effect on GPT-4 (0.70 to 0.40), when the few-shot examples are hallucination-free. We also conduct a qualitative evaluation using hallucination-free and improved training data. We find that common quantitative metrics do not correlate well with faithfulness and quality. Finally, we test GPT-4 for automatic hallucination detection, which clearly outperforms common baselines.

Automated L2 speech assessment can assign proficiency labels, but often lacks interpretability. We propose a rubric-guided SpeechLLM for multi-aspect, multi-granular assessment, trained with a hybrid objective combining supervised fine-tuning and Bounded Direct Preference Optimization. The model jointly predicts ordinal labels at the sentence-level (accuracy, fluency, prosody), word/phoneme-level accuracy, and generates a natural-language rationale in the same response. On SpeechOcean762, our approach matches or outperforms single-granularity models while remaining competitive with prior approaches. We analyze rationale reliability along two axes: self-consistency with model predictions and alignment with ground-truth labels, using sentiment consistency (plausibility) and mention-based agreement (faithfulness). Rationales are plausible at the sentence level, but faithfulness degrades at the word/phoneme level: references are sparse and weakly aligned with token-level labels.

Self-consistency (SC), leveraging multiple samples from LLMs, shows significant gains on various reasoning tasks but struggles with free-form generation due to the difficulty of aggregating answers. Its variants, UCS and USC, rely on sample selection or voting mechanisms to improve output quality. These methods, however, face limitations due to their inability to fully utilize the nuanced consensus knowledge present within multiple candidate samples, often resulting in suboptimal outputs. We propose Fine-Grained Self-Consistency (FSC) to addresses these limitations by extracting and integrating segment-level commonalities from candidate samples, enhancing the performance of LLMs both in open-ended and reasoning tasks. Based on this, we present two additional strategies: candidate filtering, which enhances overall quality by identifying highly similar candidate sets, and merging, which reduces input token requirements by combining similar samples. The effectiveness of FSC is demonstrated through extensive experiments on various tasks, including summarization, code generation, and mathematical reasoning, using GPT-3.5-turbo and GPT-4. The results indicate significant improvements over baseline methods, showcasing the potential of FSC to optimize output quality by effectively synthesizing fine-grained consensus knowledge from multiple samples.

Hallucinations in large language models (LLMs) are commonly regarded as errors to be minimized. However, recent perspectives suggest that some hallucinations may encode creative or epistemically valuable content, a dimension that remains underquantified in current literature. Existing hallucination detection methods primarily focus on factual consistency, struggling to handle heterogeneous scientific tasks and balance creativity with accuracy. To address these challenges, we propose HIC-Bench, a novel evaluation framework that categorizes hallucinations into Intelligent Hallucinations (IH) and Defective Hallucinations (DH), enabling systematic investigation of their interplay in LLM creativity. HIC-Bench features three core characteristics: (1) Structured IH/DH Assessment. using a multi-dimensional metric matrix integrating Torrance Tests of Creative Thinking (TTCT) metrics (Originality, Feasibility, Value) with hallucination-specific dimensions (scientific plausibility, factual deviation); (2) Cross-Domain Applicability. spanning ten scientific domains with open-ended innovation tasks; and (3) Dynamic Prompt Optimization. leveraging the Dynamic Hallucination Prompt (DHP) to guide models toward creative and reliable outputs. The evaluation process employs multiple LLM judges, averaging scores to mitigate bias, with human annotators verifying IH/DH classifications. Experimental results reveal a nonlinear relationship between IH and DH, demonstrating that creativity and correctness can be jointly optimized. These insights position IH as a catalyst for creativity and reveal the ability of LLM hallucinations to drive scientific innovation.Additionally, the HIC-Bench offers a valuable platform for advancing research into the creative intelligence of LLM hallucinations.

Aligning Large Language Models (LLMs) with human preferences is crucial for their deployment in real-world applications. Recent advancements in Self-Rewarding Language Models suggest that an LLM can use its internal reward models (such as LLM-as-a-Judge) yuanself to generate preference data, improving alignment performance without costly human annotation. However, we find that different internal reward models within the same LLM often generate inconsistent preferences. This inconsistency raises concerns about the reliability of self-generated preference data, hinders overall alignment performance, and highlights the need for further research to ensure reliable and coherent alignment with human preferences. To address this limitation, we propose Self-Consistent Internal Rewards (SCIR), a novel framework designed to enhance consistency among internal reward models during training. In each training step, we collect preference predictions from multiple pre-defined internal reward models and enforce consistency and confidence through an inconsistency penalty mechanism, thereby improving the reliability of these internal reward models. We selectively use data with consistent predictions for preference optimization, ensuring the quality of the preference data. By employing self-consistent internal rewards, our method significantly improves the alignment performance and reward modeling capability of LLMs, outperforming baseline methods by a notable margin.

Evaluation benchmarks are the cornerstone of measuring capabilities of large language models (LLMs), as well as driving progress in said capabilities. Originally designed to make claims about capabilities (or lack thereof) in fully pretrained models, evaluation benchmarks are now also extensively used to decide between various training choices. Despite this widespread usage, we rarely quantify the variance in our evaluation benchmarks, which dictates whether differences in performance are meaningful. Here, we define and measure a range of metrics geared towards measuring variance in evaluation benchmarks, including seed variance across initialisations, and monotonicity during training. By studying a large number of models -- both openly available and pretrained from scratch -- we provide empirical estimates for a variety of variance metrics, with considerations and recommendations for practitioners. We also evaluate the utility and tradeoffs of continuous versus discrete performance measures and explore options for better understanding and reducing this variance. We find that simple changes, such as framing choice tasks (like MMLU) as completion tasks, can often reduce variance for smaller scale (sim7B) models, while more involved methods inspired from human testing literature (such as item analysis and item response theory) struggle to meaningfully reduce variance. Overall, our work provides insights into variance in evaluation benchmarks, suggests LM-specific techniques to reduce variance, and more generally encourages practitioners to carefully factor in variance when comparing models.

LLM confidence signals are used for abstention, routing, and safety-critical decisions. No standard practice exists for checking whether a confidence signal carries item-level information before building on it. We transfer the validity screening principle from clinical personality assessment (PAI, MMPI-3) as a portable protocol for benchmark-based LLM confidence data. The protocol specifies three core indices (L, Fp, RBS), a structural indicator (TRIN), and an item-sensitivity statistic, computed from a single 2x2 contingency table. A three-tier classification system (Invalid, Indeterminate, Valid) draws on four clinical traditions. Validated on 20 frontier LLMs across 524 items, four models are classified Invalid, two Indeterminate. Valid-profile models show mean r = .18 (15/16 significant). Invalid-profile models show mean r = -.20 (d = 2.48). Cross-benchmark validation on 18 models using MMLU with verbalized confidence and on external data from Yang et al. (2024) confirms the screen transfers across benchmarks and probe formats. All data and code: https://github.com/synthiumjp/validity-scaling-llm

Probabilistic decoding in Large Language Models (LLMs) often yields inconsistent outputs, particularly on complex or long-form questions. Self-Consistency (SC) mitigates this for short-form QA by majority voting over exact strings, whereas Universal Self-Consistency (USC) and Weighted Unigram Consistency Score (WUCS) extend to long-form responses but lose accuracy on short-form benchmarks. We introduce Latent Self-Consistency (LSC), which selects the most semantically consistent response using learnable token embeddings. LSC's lightweight forward processing of summary tokens only introduces negligible runtime overhead (at most 0.9%) on top of standard decoding of the base LLM, and requires no changes to the model architecture. Across 6 short-form and 5 long-form reasoning benchmarks (e.g., MATH, MMLU, TruthfulQA), LSC surpasses SC, USC, and WUCS on both short-form and long-form on average performance, while adding negligible computational overhead on vanilla inference. These results position LSC as a reliable consistency-selection method that works effectively across various answer formats. Additionally, LSC provides well-calibrated confidence estimates, maintaining low expected calibration error across both answer formats.

Measuring complexity from a model's own code is biased by failure: a hard prompt can produce a short, broken program that looks simple. We instead score each prompt's structure on six dimensions from the prompt alone, never from generated code or test results. We draw 5,000 Python prompts across six bands of a preliminary single-rater score, and four out-of-panel LLM judges rescore every prompt (composite ICC(2,1) = 0.872). Twenty-one models answer every prompt (105,000 generations), and an independent LLM auditor, checked against known-answer cases, reviews each unit-test verdict. Averaged over models, pass rate has a nonmonotone breakpoint at composite 14.0: 83.4% at or below, 90.1% above. That jump is not a law of difficulty: with task-type fixed effects the breakpoint stays at 14.0 but the adjusted jump shrinks from 6.8 to 0.4 points, and neither of the benchmark's two prompt sources reproduces it on its own. Nineteen models pass more often above their own breakpoint and two less often; one of those declines comes from API calls that returned nothing. In post hoc follow-up, Kimi K2.5 shows the clearest decline, at 14.25, under a prompt-source control and within the later source alone. Among failed generations with measurable output complexity, 25.0% pair a prompt composite above 8 with output complexity of at most 10, twice the share among passing ones. The pooled breakpoint moves to 11.25 on raw test outcomes. Human agreement with the index is moderate, paraphrase and cross-language rescoring preserve its ordering, and overidentification tests reject the six dimensions as instruments, so we make no causal claim. The contribution is a prompt-side way to locate where each model's pass rate changes and to test how much that depends on the prompt mix and on how success is scored.

In this paper, we present a novel approach for improving the quality and consistency of generated outputs from large-scale pre-trained language models (LLMs). Self-consistency has emerged as an effective approach for prompts with fixed answers, selecting the answer with the highest number of votes. In this paper, we introduce a generalized framework for self-consistency that extends its applicability beyond problems that have fixed-answer answers. Through extensive simulations, we demonstrate that our approach consistently recovers the optimal or near-optimal generation from a set of candidates. We also propose lightweight parameter-free similarity functions that show significant and consistent improvements across code generation, autoformalization, and summarization tasks, even without access to token log probabilities. Our method incurs minimal computational overhead, requiring no auxiliary reranker models or modifications to the existing model.

Equivariance theory predicts that an architectural symmetry prior reduces sample complexity by a factor of |G|; this is widely cited but rarely measured as a scaling law with controls that separate the prior from its confounds. On a controlled C_n-symmetric task, we report three findings. First, a wrong-group control with identical orbit size and matched compute is worse than no constraint (joint pairwise CI [+0.79, +3.26] excludes zero, robust across estimators); misaligned constraint is actively harmful, not merely unhelpful. Second, an augmentation baseline equipped with test-time orbit averaging matches the equivariant model exactly -- bit-identical per-epoch validation curves across matched cells -- so the architecture-vs-augmentation gap is conditional on asymmetric test-time computation, not unconditional. Third, the relative exchange rate beta_diff = 1.28 is consistent in sign and order of magnitude with the theoretical 1.0 (single-level CI [+0.92, +2.05]); the more conservative two-level bootstrap (seeds x group sizes) widens this to [-0.63, +1.72], including zero, and a finer-N replication on a sqrt(2)-spaced grid is inconclusive (point estimate -0.82). The methodological contributions -- the relative-rate estimator that cancels the shared-difficulty confound, the wrong-group control, and a pre-specified failure taxonomy -- transfer to any inductive bias whose strength can be parameterised. Honest scoping: the primary estimator beta_diff was adopted post-hoc after the initial analysis revealed a positive-slope identifiability problem; the design was never externally pre-registered; and the headline number rests on an OLS slope over seven group sizes on a coarse N grid. This is an exploratory study, not a confirmatory measurement; the wrong-group result is the cleanest finding and the one we report with the most confidence. A registered replication on fresh seeds is future work.

LLM-as-a-Judge has been widely applied to evaluate and compare different LLM alignmnet approaches (e.g., RLHF and DPO). However, concerns regarding its reliability have emerged, due to LLM judges' biases and inconsistent decision-making. Previous research has developed evaluation frameworks to assess reliability of LLM judges and their alignment with human preferences. However, the employed evaluation metrics often lack adequate explainability and fail to address LLM internal inconsistency. Additionally, existing studies inadequately explore the impact of various prompt templates when applying LLM-as-a-Judge methods, leading to potentially inconsistent comparisons between different alignment algorithms. In this work, we systematically evaluate LLM-as-a-Judge on alignment tasks by defining more theoretically interpretable evaluation metrics and explicitly mitigating LLM internal inconsistency from reliability metrics. We develop an open-source framework to evaluate, compare, and visualize the reliability and alignment of LLM judges, which facilitates practitioners to choose LLM judges for alignment tasks. In the experiments, we examine effects of diverse prompt templates on LLM-judge reliability and also demonstrate our developed framework by comparing various LLM judges on two common alignment datasets (i.e., TL;DR Summarization and HH-RLHF-Helpfulness). Our results indicate a significant impact of prompt templates on LLM judge performance, as well as a mediocre alignment level between the tested LLM judges and human evaluators.

A benchmark is not merely a collection of test questions; it naturally induces an empirical measure over difficulty. In this work, we develop this measure-based view of benchmark evaluation. This perspective allows us to identify and formalize two central failure modes: measure mismatch, where the benchmark difficulty distribution differs from the intended difficulty distribution, and coverage gaps, where uncovered regions along the difficulty axis prevent benchmarks from detecting meaningful localized training progress. We propose two diagnostic metrics that capture these two failure modes and provide rigorous theoretical guarantees. Empirically, we analyze 25 popular benchmarks spanning 7 domains using carefully estimated item difficulties. Our results show that many benchmarks implicitly overweight certain difficulty regions, so their leaderboard conclusions should be interpreted with caution; others leave substantial portions of the difficulty axis uncovered, limiting their ability to monitor the model improvement process. Finally, we turn diagnosis into action by introducing BenchFill, a simple difficulty-targeted item rewriting algorithm that effectively reduces local coverage gaps.

As large language models are increasingly deployed to simulate diverse human characters, ensuring persona fidelity, defined as the extent to which an agent's behavior consistently reflects the psychological and stylistic characteristics of a target persona, has become a critical requirement. However, existing evaluation paradigms primarily rely on either holistic LLM-based judges, which are prone to "holistic appraisal hallucination'', or static psychometric inventories, which fail to capture the context-dependent fidelity required in dynamic dialogue. To address these limitations, we propose PRISM (Persona Reasoning with Inverse SFL-based Modeling), a psycholinguistically grounded framework that reformulates persona fidelity evaluation as a structured inverse inference task. Inspired by Systemic Functional Linguistics (SFL), PRISM decomposes persona fidelity into three functional dimensions: Task Framing, Interpersonal Stance, and Linguistic Style. It estimates dimension-specific evidence over a persona-conditioned label space and aggregates these signals into an interpretable and auditable evaluation process. Experiments show that PRISM yields more accurate and stable judgements than traditional holistic judging, providing a more reliable framework for persona fidelity evaluation.

Recent benchmarks rely on user simulators to evaluate AI agents in multi-turn interaction. While existing simulation techniques demonstrate surface fidelity to human style and behavior, ecologically valid interactive benchmarking also requires alignment in when and how agents fail across simulated and real user populations. We find that existing simulators lack outcome calibration: agreement with observed success rates and failure patterns when real users interact with the same agent. We introduce Calibrated User Embeddings (CUE), a framework that both encodes observed sessions and samples continuous representations, then decodes them into persona commands to steer LLMs to act as user simulators without training. Through this, we evaluate user-conditioned replay of past sessions and aggregate metric agreement when sampling novel personas for the same tasks. On τ^2-Bench, CUEd simulators commit fewer simulator-attributed errors and more faithfully reproduce real-user agent failure modes, aggregate success rates, and outcomes for specific task-user pairs than other persona-based simulation methods. These gains coexist with competitive user fidelity as measured using metrics established in prior work. After being fit to mostly customer support interactions, the same CUEd simulators generalize to document creation, math tutoring, and casual conversation, and remain effective across different simulator LLMs without CUE retraining.

Large language model (LLM) judges are often used as rewards to train policies on objectives that deterministic verifiers cannot capture. However, these rewards are often collapsed to pass/fail ({0, 1}), which reports the verdict but not how well a response met each criterion. We model each pass/fail verdict as a score on an unreported scale, compared with one cutoff. A stretch of that scale moves every score proportionally toward or away from the cutoff, but never across it, so no verdict changes. A policy is therefore free to apply any stretch without changing anything the panel reports. Under a joint-Gaussian model, a third grade adds a second threshold and removes this affine stretch ambiguity. On MATH and SciBench outputs from one seven-criterion judge, all 14 constructed criterionwise stretches were invisible after binarization but visible with three grades. At n=1{,}024, a test given both population laws had at least 96.5% power at a 1.5times stress. Retaining grades closes one blind spot created by binarization, but verdicts alone remain insufficient as some changes are still indistinguishable from genuine improvement. These include arbitrary within-grade changes and fixed-covariance, loading-aligned mean shifts -- the signature of a sycophancy-shaped lift the panel reads as competence. The shared-factor reference approximation fit MATH and SciBench but not HealthBench, delineating its empirical scope. We recommend keeping at least three grades (for example, asking the judge whether each criterion is fully, partially, or not met and rewarding {0, 0.5, 1}), and externally validating gains along the remaining direction, which no finer scale removes.

We conduct a systematic audit of three widely used reasoning benchmarks, SocialIQa, FauxPas-EAI, and ToMi, and uncover pervasive flaws in both benchmark items and evaluation methodology. Using five LLMs (GPT-{3, 3.5, 4, o1}, and LLaMA 3.1) as diagnostic tools, we identify structural, semantic, and pragmatic issues in benchmark design (e.g., duplicated items, ambiguous wording, and implausible answers), as well as scoring procedures that prioritize output form over reasoning process. Through systematic human annotation and re-evaluation on cleaned benchmark subsets, we find that model scores often improve not due to due to erratic surface wording variations and not to improved reasoning. Infact, further analyses show that model performance is highly sensitive to minor input variations such as context availability and phrasing, revealing that high scores may reflect alignment with format-specific cues rather than consistent inference based on the input. These findings challenge the validity of current benchmark-based claims about reasoning in LLMs, and highlight the need for evaluation protocols that assess reasoning as a process of drawing inference from available information, rather than as static output selection. We release audited data and evaluation tools to support more interpretable and diagnostic assessments of model reasoning.

Large Language Model (LLM) personas with explicit specifications of attributes, background, and behavioural tendencies are increasingly used to simulate human conversations for tasks such as user modeling, social reasoning, and behavioural analysis. Ensuring that persona-grounded simulations faithfully reflect human conversational behaviour is therefore critical. However, current evaluation practices largely rely on LLM-as-a-judge approaches, offering limited grounding in observable human behavior and producing opaque scalar scores. We address this gap by proposing Eval4Sim, an evaluation framework that measures how closely simulated conversations align with human conversational patterns across three complementary dimensions. Adherence captures how effectively persona backgrounds are implicitly encoded in generated utterances, assessed via dense retrieval with speaker-aware representations. Consistency evaluates whether a persona maintains a distinguishable identity across conversations, computed through authorship verification. Naturalness reflects whether conversations exhibit human-like flow rather than overly rigid or optimized structure, quantified through distributions derived from dialogue-focused Natural Language Inference. Unlike absolute or optimization-oriented metrics, Eval4Sim uses a human conversational corpus (i.e., PersonaChat) as a reference baseline and penalizes deviations in both directions, distinguishing insufficient persona encoding from over-optimized, unnatural behaviour. Although demonstrated on PersonaChat, the applicability of Eval4Sim extends to any conversational corpus containing speaker-level annotations.

Ensuring alignment, which refers to making models behave in accordance with human intentions [1,2], has become a critical task before deploying large language models (LLMs) in real-world applications. For instance, OpenAI devoted six months to iteratively aligning GPT-4 before its release [3]. However, a major challenge faced by practitioners is the lack of clear guidance on evaluating whether LLM outputs align with social norms, values, and regulations. This obstacle hinders systematic iteration and deployment of LLMs. To address this issue, this paper presents a comprehensive survey of key dimensions that are crucial to consider when assessing LLM trustworthiness. The survey covers seven major categories of LLM trustworthiness: reliability, safety, fairness, resistance to misuse, explainability and reasoning, adherence to social norms, and robustness. Each major category is further divided into several sub-categories, resulting in a total of 29 sub-categories. Additionally, a subset of 8 sub-categories is selected for further investigation, where corresponding measurement studies are designed and conducted on several widely-used LLMs. The measurement results indicate that, in general, more aligned models tend to perform better in terms of overall trustworthiness. However, the effectiveness of alignment varies across the different trustworthiness categories considered. This highlights the importance of conducting more fine-grained analyses, testing, and making continuous improvements on LLM alignment. By shedding light on these key dimensions of LLM trustworthiness, this paper aims to provide valuable insights and guidance to practitioners in the field. Understanding and addressing these concerns will be crucial in achieving reliable and ethically sound deployment of LLMs in various applications.

LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated. Existing benchmarks use hand-authored scenarios and prompted simulators, aggregate empathy into one score, and overlook judge biases such as same-family favoritism and scale drift. We introduce CompanionBench, an interactive bilingual benchmark. To our knowledge, it is the first companion benchmark to ground both its scenarios and a trained user simulator in de-identified real-world data. A hidden disclosure gate branches each persona's trajectory on the agent's own behavior, controlling the interaction state space without scripting dialogue. We operationalize ten capabilities derived from 25 theories across psychology and counseling, four of them not graded explicitly by prior work: holding ambiguity, selfobject responsiveness, positive resonance and calibrated challenge. Agents are assessed on two complementary axes: a subjective ten-capability rubric and a deterministic measure of whether deeper disclosure was earned. A cross-family panel dilutes same-family favoritism; an Item Response Theory model separates agent quality from judge severity. Theory fixes what to measure and how personas are structured; real data supply events, history, and profiles -- coverage from theory, authenticity from data. Rankings are reproducible in both languages (rho = 0.996 ZH / 0.953 EN). Evaluating 28 agents reveals capability-level differences obscured by aggregate scores. Emotion regulation and calibrated challenge remain common weaknesses; holding ambiguity discriminates most. Role-play agents rank near the bottom: immersion does not imply relational competence. Across agents, the dominant failure mode is substituting surface warmth for substantive relational support. We will release 500 Chinese-English parallel pairs and the evaluation code.

Automatically generated questions often suffer from problems such as unclear expression or factual inaccuracies, requiring a reliable and comprehensive evaluation of their quality. Human evaluation is frequently used in the field of question generation (QG) and is one of the most accurate evaluation methods. It also serves as the standard for automatic metrics. However, there is a lack of unified evaluation criteria, which hampers the development of both QG technologies and automatic evaluation methods. To address this, we propose QGEval, a multi-dimensional Evaluation benchmark for Question Generation, which evaluates both generated questions and existing automatic metrics across 7 dimensions: fluency, clarity, conciseness, relevance, consistency, answerability, and answer consistency. We demonstrate the appropriateness of these dimensions by examining their correlations and distinctions. Analysis with QGEval reveals that 1) most QG models perform unsatisfactorily in terms of answerability and answer consistency, and 2) existing metrics fail to align well with human assessments when evaluating generated questions across the 7 dimensions. We expect this work to foster the development of both QG technologies and automatic metrics for QG.