LLM Evaluation

LLM: Large Language Model

Follow this topic Ask about this topic

The field studies how to measure language-model capabilities, reliability, and risks, and whether those measurements support sound comparisons and deployment decisions. Approaches include curated and synthetic benchmarks, psychometric and adaptive testing, execution-based checks, robustness tests, and human or LLM-based judges. Researchers also examine calibration, uncertainty, contamination, prompt sensitivity, and whether scores reflect the intended capability across languages and domains. A central tension is that efficient aggregate scores can hide important errors and measurement biases, while more realistic, human-grounded evaluations are costly and difficult to standardize.

Latest papers

2,432 papers286 papers in the last four weeks, up 16% on the four weeks before. 1.7% of all new papers.

Mar 18, 2026

Can LLMs Reason Like Automated Theorem Provers for Rust Verification? VCoT-Bench: Evaluating via Verification Chain of Thought

As Large Language Models (LLMs) increasingly assist secure software development, their ability to meet the rigorous demands of Rust program verification remains unclear. Existing evaluations treat Rust verification as a black box, assessing models only by binary pass or fail outcomes for proof hints. This obscures whether models truly understand the logical deductions required for verifying nontrivial Rust code. To bridge this gap, we introduce VCoT-Lift, a framework that lifts low-level solver reasoning into high-level, human-readable verification steps. By exposing solver-level reasoning as an explicit Verification Chain-of-Thought, VCoT-Lift provides a concrete ground truth for fine-grained evaluation. Leveraging VCoT-Lift, we introduce VCoT-Bench, a comprehensive benchmark of 1,988 VCoT completion tasks for rigorously evaluating LLMs' understanding of the entire verification process. VCoT-Bench measures performance along three orthogonal dimensions: robustness to varying degrees of missing proofs, competence across different proof types, and sensitivity to the proof locations. Evaluation of ten state-of-the-art models reveals severe fragility, indicating that current LLMs fall well short of the reasoning capabilities exhibited by automated theorem provers.

Zichen Xie, Wenxi Wang

University of Virginia.

Formal Verification LLM Evaluation CoT Reasoning

Mar 18, 2026

GRAFITE: Generative Regression Analysis Framework for Issue Tracking and Evaluation

Large language models (LLMs) are largely motivated by their performance on popular topics and benchmarks at the time of their release. However, over time, contamination occurs due to significant exposure of benchmark data during training. This poses a risk of model performance inflation if testing is not carefully executed. To address this challenge, we present GRAFITE, a continuous LLM evaluation platform through a comprehensive system for maintaining and evaluating model issues. Our approach enables building a repository of model problems based on user feedback over time and offers a pipeline for assessing LLMs against these issues through quality assurance (QA) tests using LLM-as-a-judge. The platform enables side-by-side comparison of multiple models, facilitating regression detection across different releases. The platform is available at https://github.com/IBM/grafite. The demo video is available at www.youtube.com/watch?v=XFZyoleN56k.

Ja Young Lee, Mírian Silva, Mohamed Nasr +6

IBM Research - AI

LLM Evaluation Data Contamination in Language Models

Mar 18, 2026

Toward Measuring Structural Drift in LLM Communication Loops

Large language models increasingly run in stateful pipelines that assemble each prompt from retrieval, memory, tools, and other agents. Such pipelines drift: information that should shape the next response is dropped, compressed, or misrouted while every component still reports success. Existing diagnostics miss this because they evaluate isolated prompts, responses, or task scores, whereas what decouples is the relation between a prompt and the response it draws. Here we show that treating the prompt to response to next prompt chain as the fundamental unit of analysis makes these relations measurable. We introduce structural communication coherence, quantified by two metrics: communication closure, which asks if what the pipeline returns at one turn matches what it faces next, and normalized conditional action contribution, which measures how much a sent message resolves the subsequent reply. Across 2,171 human to human, 58 human to LLM, and 8 LLM to LLM dialogues, these metrics reveal directional interaction structures; crucially, the measured contribution drops by 87 to 92% when a response is swapped for one from another turn, leaving surrounding prompts untouched. Because this approach requires no labels, healthy reference data, or predefined rules only the raw prompts and responses drift can be defined and measured directly from operational traffic, rather than inferred from eventual task failure. Establishing prospective detection performance is the next step.

Wael Hafez, Amir Nazeri, Chenan Wei

Semarx Research LLC Alexandria, VA, USA

Mar 18, 2026

Event-Centric Human Value Understanding in News-Domain Texts: An Actor-Conditioned, Multi-Granularity Benchmark

Existing human value datasets do not directly support value understanding in factual news: many are actor-agnostic, rely on isolated utterances or synthetic scenarios, and lack explicit event structure or value direction. We present NEVU (News Event-centric Value Understanding), a benchmark for actor-conditioned, event-centric, and direction-aware human value recognition in factual news. NEVU evaluates whether models can identify value cues, attribute them to the correct actor, and determine value direction from grounded evidence. Built from 2{,}865 English news articles, NEVU organizes annotations at four semantic unit levels (Subevent, behavior-based composite event, story-based composite event, and Article) and labels \mbox{(unit, actor)} pairs for fine-grained evaluation across local and composite contexts. The annotations are produced through an LLM-assisted pipeline with staged verification and targeted human auditing. Using a hierarchical value space with 54 fine-grained values and 20 coarse-grained categories, NEVU covers 45{,}793 unit--actor pairs and 168{,}061 directed value instances. We provide unified baselines for proprietary and open-source LLMs, and find that lightweight adaptation (LoRA) consistently improves open-source models, showing that although NEVU is designed primarily as a benchmark, it also supports supervised adaptation beyond prompting-only evaluation. Data availability is described in Appendix~\ref{app:data_code_availability}.

Yao Wang, Xin Liu, Zhuochen Liu +5

Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tsukuba, Japan · Intelligent Platforms Research Institute, National Institute of Advanced Industrial Science and Technology, Tokyo, Japan · Digital Science Center, University of Innsbruck, Innsbruck, Austria +2

Mar 18, 2026

From Isolated Scoring to Collaborative Ranking: A Comparison-Native Framework for LLM-Based Paper Evaluation

Large language models (LLMs) are currently applied to scientific paper evaluation by assigning an absolute score to each paper independently. However, since score scales vary across conferences, time periods, and evaluation criteria, models trained on absolute scores are prone to fitting narrow, context-specific rules rather than developing robust scholarly judgment. To overcome this limitation, we propose shifting paper evaluation from isolated scoring to collaborative ranking. In particular, we design a

\textbf{C}

omparison-

\textbf{N}

ative framework for

\textbf{P}

aper

\textbf{E}

valuation (

\textbf{CNPE}

), integrating comparison into both data construction and model learning. We first propose a graph-based similarity ranking algorithm to facilitate the sampling of more informative and discriminative paper pairs from a collection. We then enhance relative quality judgment through supervised fine-tuning and reinforcement learning with comparison-based rewards. At inference, the model performs pairwise comparisons over sampled paper pairs and aggregates these preference signals into a global relative quality ranking. Experimental results demonstrate that our framework achieves an average relative improvement of 21.8% over the strong baseline DeepReview-14B, while exhibiting robust generalization to five previously unseen datasets. Our code is available at https://github.com/ECNU-Text-Computing/ComparisonReview.

Pujun Zheng, Jiacheng Yao, Jinquan Zheng +6

School of Economics and Management, East China Normal University · School of Information Management, Wuhan University · China Academic Degrees & Graduate Education Development Center

Automated Peer Review Scholarly Peer Review Learning to Rank Pairwise Preference Learning LLM Evaluation

Mar 17, 2026

REAL: Regression-Aware Reinforcement Learning for LLM-as-a-Judge

Large language models (LLMs) are increasingly deployed as automated evaluators that assign numeric scores to model outputs, a paradigm known as LLM-as-a-Judge. However, standard Reinforcement Learning (RL) methods typically rely on binary rewards (e.g., 0-1 accuracy), thereby ignoring the ordinal structure inherent in regression tasks; for instance, they fail to recognize that predicting 4 is significantly better than predicting 1 when the ground truth is 5. Conversely, existing regression-aware approaches are often confined to Supervised Fine-Tuning (SFT), limiting their ability to explore optimal reasoning paths. To bridge this gap, we propose REAL (REgression-Aware Reinforcement Learning), a principled RL framework designed to optimize regression rewards, and also proven to be optimal for correlation metrics. A key technical challenge is that the regression objective is explicitly policy-dependent, thus invalidating standard policy gradient methods. To address this, we employ the generalized policy gradient estimator, which naturally decomposes optimization into two complementary components: (1) exploration over Chain-of-Thought (CoT) trajectory, and (2) regression-aware prediction refinement of the final score. Extensive experiments across model scales (8B to 32B) demonstrate that REAL consistently outperforms both regression-aware SFT baselines and standard RL methods, exhibiting significantly better generalization on out-of-domain benchmarks. On Qwen3-32B specifically, we achieve gains of +8.40 Pearson and +7.20 Spearman correlation over the SFT baseline, and +18.30/+11.20 over the base model. These findings highlight the critical value of integrating regression objectives into RL exploration for accurate LLM evaluation.

Yasi Zhang, Tianyu Chen, Mingyuan Zhou +3

University of California, Los Angeles · The University of Texas at Austin · Google Research

RL for Language Models Reinforcement Learning LLM-as-a-Judge LLM Evaluation

Mar 17, 2026

Evaluating LLM-Simulated Conversations in Modeling Inconsistent and Uncollaborative Behaviors in Human Social Interaction

Simulating human conversations using large language models (LLMs) has emerged as a scalable methodology for modeling human social interaction. This paper reconsiders the evaluation of simulated conversations by explicitly recognizing that human conversations inherently involve inconsistent and uncollaborative behaviors, such as misunderstandings and interruptions. Since these behaviors contribute to the complexity of human social interaction, we argue that LLM-simulated conversations should reproduce them at frequencies comparable to those observed in human conversations. To support a detailed and interpretable evaluation of these behaviors, we introduce CoCoEval, a framework consisting of an evaluation scheme based on turn-level detection of 10 types of inconsistent and uncollaborative behaviors and a benchmark for simulating conversations in professional scenarios involving collaboration and conflict. Using CoCoEval, we compare human conversations with those simulated by GPT-4.1, GPT-5.1, and Claude Opus 4. The results show that (1) LLM-simulated conversations exhibit far fewer inconsistent and uncollaborative behaviors than human conversations under vanilla prompting, and (2) prompt engineering and supervised fine-tuning do not provide reliable control over these behaviors, often leading to the overproduction of specific behaviors. CoCoEval identifies gaps between human and LLM-simulated conversations that are not captured by conventional evaluation based on conversation-level Likert scales, raising concerns about the use of LLMs as proxies for human social interaction.

Ryo Kamoi, Ameya Godbole, Binglin Zhou +5

Penn State University · University of Southern California · Microsoft Corporation

LLM Evaluation Human Behavior Simulation Large Language Model-Based Social Simulation

Mar 17, 2026

Evaluating Ill-Defined Tasks in Large Language Models

Many evaluations of Large Language Models (LLMs) target tasks that are inherently ill-defined, with unclear input and output spaces and ambiguous success criteria. We analyze why existing evaluation benchmarks and metrics fail to provide reliable or diagnostic signals of model capability for such tasks. We examine two case studies: Complex Instruction Following (CIF), where we identify recurring issues including limited coverage of real-world instruction complexity, sensitivity to instruction phrasing, inconsistent and non-comparable metrics, and instability introduced by LLM-based judges; and Natural Language to Mermaid Sequence Diagrams (NL2Mermaid), where we show how multi-faceted evaluation criteria can yield actionable insights beyond aggregate scores. Together, these case studies show that current evaluations frequently conflate distinct failure modes, yielding scores that are unstable, non-diagnostic, and difficult to act upon. Our findings expose fundamental limitations in existing evaluation practices for ill-defined tasks and motivate more robust, interpretable evaluation designs.

Yi Zhou, Basel Shbita

IBM Research, San Jose, CA

LLM Evaluation Instruction Following

Mar 17, 2026

LLM NL2SQL Robustness: Surface Noise vs. Linguistic Variation in Traditional and Agentic Settings

Robustness evaluation for Natural Language to SQL (NL2SQL) systems is essential because real-world database environments are dynamic, noisy, and continuously evolving, whereas conventional benchmark evaluations typically assume static schemas and well-formed user inputs. In this work, we introduce a robustness evaluation benchmark containing approximately ten types of perturbations and conduct evaluations under both traditional and agentic settings. We assess multiple state-of-the-art large language models (LLMs), including Grok-4.1, Gemini-3-Pro, Claude-Opus-4.6, and GPT-5.2. Our results show that these models generally maintain strong performance under several perturbations; however, notable performance degradation is observed for surface-level noise (e.g., character-level corruption) and linguistic variation that preserves semantics while altering lexical or syntactic forms. Furthermore, we observe that surface-level noise causes larger performance drops in traditional pipelines, whereas linguistic variation presents greater challenges in agentic settings. These findings highlight the remaining challenges in achieving robust NL2SQL systems, particularly in handling linguistic variability.

Lifu Tu, Rongguang Wang, Tao Sheng +2

Oracle AI

Text-to-SQL Language Model Robustness LLM Evaluation

Mar 17, 2026

Me, Myself, and $π$ : Evaluating and Explaining LLM Introspection

A hallmark of human intelligence is Introspection-the ability to assess and reason about one's own cognitive processes. Introspection has emerged as a promising but contested capability in large language models (LLMs). However, current evaluations often fail to distinguish genuine meta-cognition from the mere application of general world knowledge or text-based self-simulation. In this work, we propose a principled taxonomy that formalizes introspection as the latent computation of specific operators over a model's policy and parameters. To isolate the components of generalized introspection, we present Introspect-Bench, a multifaceted evaluation suite designed for rigorous capability testing. Our results show that frontier models exhibit privileged access to their own policies, outperforming peer models in predicting their own behavior. Furthermore, we provide causal, mechanistic evidence explaining both how LLMs learn to introspect without explicit training, and how the mechanism of introspection emerges via attention diffusion.

Atharv Naphade, Samarth Bhargav, Sean Lim +1

Carnegie Mellon University

Language Model Introspection Self-Reflection in Language Model Agents Metacognition in Language Models LLM Evaluation LLM Interpretability Mechanistic Interpretability

Mar 17, 2026

Arabic Morphosyntactic Tagging and Dependency Parsing with Large Language Models

LLMs perform strongly across NLP, but their ability to produce explicit grammatical analyses remains unclear. Arabic provides a challenging testbed due to its rich morphology and orthographic ambiguity, which create strong morphology-syntax interactions. We present a unified evaluation of LLMs on Arabic morphosyntactic tagging and dependency parsing, covering pre-tokenized, raw-text, and cascaded settings. We compare zero-shot prompting with retrieval-based in-context learning. Relevant demonstrations substantially improve performance. The strongest LLMs approach supervised tagging and parsing systems; however, they require substantial annotated data for demonstration retrieval and considerable computational resources. We make all code and data used in this paper publicly available.

Mohamed Adel, Bashar Alhafni, Nizar Habash

Computational Approaches to Modeling Language Lab 1New York University Abu Dhabi · 2Mohamed bin Zayed University of Artificial Intelligence

LLM Evaluation Arabic NLP Morphosyntactic Tagging In-Context Learning Dependency Parsing

Mar 17, 2026

LLMs learn scientific taste from institutional traces across the social sciences

Reinforcement-learned reasoning has powered recent AI leaps on verifiable tasks, including mathematics, code, and structure prediction. The harder bottleneck is evaluative judgment in low-verifiability domains, where no oracle anchors reward and the core question is which untested ideas deserve attention. We test whether institutional traces, the record of what fields published, where, and at which tier, can serve as a training signal for AI evaluators. Across eight social science disciplines (psychology, economics, communication, sociology, political science, management, business and finance, public administration), we built held-out four-tier research-pitch benchmarks and supervised-fine-tuned (SFT) LLMs on field-specific publication outcomes. The fine-tuned models cleared the 25 percent chance baseline and exceeded frontier-model performance by wide margins, with best single-model accuracy ranging from 55.0 percent in public administration to 85.5 percent in psychology. In management, evaluated against 48 expert gatekeepers, 174 junior researchers, and 11 frontier reasoning models, the best single fine-tuned model (Qwen3-4B) reached 59.2 percent, 17.6 percentage points above expert majority vote (41.6 percent, non-tied) and 28.1 percentage points above the frontier mean (31.1 percent). The fine-tuned models also showed calibrated confidence: confidence rose when predictions were correct and fell when wrong, mirroring how a skilled reviewer can say "I'm sure" versus "I'm guessing." Selective triage on this signal reached very high accuracy on the highest-confidence subsets in every field. Institutional traces, we conclude, encode a scalable training signal for the low-verifiability judgment on which science depends.

Ziqin Gong, Ning Li, Huaikang Zhou

School of Economics and Management, Tsinghua University, Beijing, China

Mar 17, 2026

Omanic: Towards Step-wise Evaluation of Multi-hop Reasoning in Large Language Models

Evaluating the reasoning abilities of large language models (LLMs) solely from final answers can obscure failures in intermediate steps, especially in multi-hop QA benchmarks without step-level annotations. To address this gap, we introduce Omanic, an open-domain 4-hop QA benchmark designed not only to measure final-answer accuracy but also to diagnose where reasoning breaks down. Omanic contains 10,296 machine-generated training examples (OmanicSynth) and 967 expert-reviewed human-annotated evaluation examples (OmanicBench), with each evaluation question decomposed into single-hop sub-questions, intermediate answers, and structured graph topologies. Experiments with proprietary and open-source LLMs show that Omanic is challenging, while step-wise analysis reveals a later-hop bottleneck, factual knowledge floor, and error propagation along reasoning chains. Fine-tuning on OmanicSynth transfers to six reasoning and mathematics benchmarks, yielding a 7.41-point average gain and validating its effectiveness as supervision for reasoning-capability transfer. We release the data at https://huggingface.co/datasets/li-lab/Omanic and the code at https://github.com/XiaojieGu/Omanic.

Xiaojie Gu, Sherry T. Tong, Aosong Feng +8

The University of Tokyo · Yale University · Stanford University +3

LLM Evaluation Multi-Hop Reasoning Multi-Hop QA

Mar 17, 2026

BenchPreS: A Benchmark for Context-Aware Personalized Preference Selectivity of Persistent-Memory LLMs

Large language models (LLMs) increasingly store user preferences in persistent memory to support personalization across interactions. However, in third-party communication settings governed by social and institutional norms, some user preferences may be inappropriate to apply. We introduce BenchPreS, which evaluates whether memory-based user preferences are appropriately applied or suppressed across communication contexts. Using two complementary metrics, Misapplication Rate (MR) and Appropriate Application Rate (AAR), we find even frontier LLMs struggle to apply preferences in a context-sensitive manner. Models with stronger preference adherence exhibit higher rates of over-application, and neither reasoning capability nor prompt-based defenses fully resolve this issue. These results suggest current LLMs treat personalized preferences as globally enforceable rules rather than as context-dependent normative signals.

Sangyeon Yoon, Hyesoo Hong, Wonje Jeung +5

Yonsei University · LG AI Research

Personalized Language Models LLM Evaluation Persistent Memory for Language Models User Preference Modeling Normative Conformity in Language Models

Mar 17, 2026

How often do Answers Change? Estimating Recency Requirements in Question Answering

Large language models (LLMs) often rely on outdated knowledge when answering time-sensitive questions, leading to confident yet incorrect responses. Without explicit signals indicating whether up-to-date information is required, models struggle to decide when to retrieve external evidence, how to reason about stale facts, and how to rank answers by their validity. Existing benchmarks either periodically refresh answers or rely on fixed templates, but they do not reflect on how frequently answers change or whether a question inherently requires up-to-date information. To address this gap, we introduce a recency-stationarity taxonomy that categorizes questions by how often their answers change and whether this change frequency is time-invariant or context-dependent. Building on this taxonomy, we present RecencyQA, a dataset of 4,031 open-domain questions annotated with recency and stationarity labels. Through human evaluation and empirical analysis, we show that non-stationary questions, i.e., those where context changes the recency requirement, are significantly more challenging for LLMs, with difficulty increasing as update frequency rises. By explicitly modeling recency and context dependence, RecencyQA enables fine-grained benchmarking and analysis of temporal reasoning beyond binary notions of freshness, and provides a foundation for developing recency-aware and context-sensitive question answering systems.

Bhawna Piryani, Zehra Mert, Adam Jatowt

University of Innsbruck Innsbruck, Austria · MEF University Istanbul, Turkey

Temporal QA Temporal Reasoning in Language Models LLM Evaluation Knowledge-Intensive QA

Mar 17, 2026

Who Benchmarks the Benchmarks? A Case Study of LLM Evaluation in Icelandic

This paper evaluates current Large Language Model (LLM) benchmarking for Icelandic, identifies problems, and calls for improved evaluation methods in low/medium-resource languages in particular. We show that benchmarks that include synthetic or machine-translated data that have not been verified in any way, commonly contain severely flawed test examples that are likely to skew the results and undermine the tests' validity. We warn against the use of such methods without verification in low/medium-resource settings as the translation quality can, at best, only be as good as MT quality for a given language at any given time. Indeed, the results of our quantitative error analysis on existing benchmarks for Icelandic show clear differences between human-authored/-translated benchmarks vs. synthetic or machine-translated benchmarks.

Finnur Ágúst Ingimundarson, Steinunn Rut Friðriksdóttir, Bjarki Ármannsson +2

University of Zürich · University of Iceland · The Árni Magnússon Institute for Icelandic Studies

LLM Evaluation Benchmark Validity Low-Resource Language Processing Synthetic Data Evaluation

Mar 17, 2026

More Rounds, More Noise: Why Multi-Turn Review Fails to Improve Cross-Context Verification

Cross-Context Review (CCR) improves LLM verification by separating production and review into independent sessions. A natural extension is multi-turn review: letting the reviewer ask follow-up questions, receive author responses, and review again. We call this Dynamic Cross-Context Review (D-CCR). In a controlled experiment with 30 artifacts and 150 injected errors, we tested four D-CCR variants against the single-pass CCR baseline. Single-pass CCR (F1 = 0.376) significantly outperformed all multi-turn variants, including D-CCR-2b with question-and-answer exchange (F1 = 0.303,

p < 0.001

,

d = -0.59

). Multi-turn review increased recall (+0.08) but generated 62% more false positives (8.5 vs. 5.2), collapsing precision from 0.30 to 0.20. Two mechanisms drive this degradation: (1) false positive pressure -- reviewers in later rounds fabricate findings when the artifact's real errors have been exhausted, and (2) Review Target Drift -- reviewers provided with prior Q&A exchanges shift from reviewing the artifact to critiquing the conversation itself. Independent re-review without prior context (D-CCR-2c) performed worst (F1 = 0.263), confirming that mere repetition degrades rather than helps. The degradation stems from false positive pressure in additional rounds, not from information amount -- within multi-turn conditions, more information actually helps (D-CCR-2b > D-CCR-2a). The problem is not what the reviewer sees, but that reviewing again invites noise.

Song Tae-Eun

Daejeon Jungang Cheonggua Co., Ltd.

Automated Peer Review LLM Evaluation

Mar 17, 2026

Are Large Language Models Truly Smarter Than Humans?

Public leaderboards increasingly suggest that large language models (LLMs) surpass human experts on benchmarks spanning academic knowledge, law, and programming. Yet most benchmarks are fully public, their questions widely mirrored across the internet, creating systematic risk that models were trained on the very data used to evaluate them. This paper presents three complementary experiments forming a rigorous multi-method contamination audit of six frontier LLMs: GPT-4o, GPT-4o-mini, DeepSeek-R1, DeepSeek-V3, Llama-3.3-70B, and Qwen3-235B. Experiment 1 applies a lexical contamination detection pipeline to 513 MMLU questions across all 57 subjects, finding an overall contamination rate of 13.8% (18.1% in STEM, up to 66.7% in Philosophy) and estimated performance gains of +0.030 to +0.054 accuracy points by category. Experiment 2 applies a paraphrase and indirect-reference diagnostic to 100 MMLU questions, finding accuracy drops by an average of 7.0 percentage points under indirect reference, rising to 19.8 pp in both Law and Ethics. Experiment 3 applies TS-Guessing behavioral probes to all 513 questions and all six models, finding that 72.5% trigger memorization signals far above chance, with DeepSeek-R1 displaying a distributed memorization signature (76.6% partial reconstruction, 0% verbatim recall) that explains its anomalous Experiment 2 profile. All three experiments converge on the same contamination ranking: STEM > Professional > Social Sciences > Humanities.

Eshwar Reddy M, Sourav Karmakar

Applied AI Scientist, Health Vectors · Senior AI Scientist, Intuit India

Data Contamination in Language Models LLM Evaluation LLM Auditing Memorization in Language Models

Mar 17, 2026

SciZoom: A Large-scale Benchmark for Hierarchical Scientific Summarization across the LLM Era

The explosive growth of AI research has created unprecedented information overload, increasing the demand for scientific summarization at multiple levels of granularity beyond traditional abstracts. While LLMs are increasingly adopted for summarization, existing benchmarks remain limited in scale, target only a single granularity, and predate the LLM era. Moreover, since the release of ChatGPT in November 2022, researchers have rapidly adopted LLMs for drafting manuscripts themselves, fundamentally transforming scientific writing, yet no resource exists to analyze how this writing has evolved. To bridge these gaps, we introduce SciZoom, a benchmark comprising 44,946 papers from four top-tier ML venues (NeurIPS, ICLR, ICML, EMNLP) spanning 2020 to 2025, explicitly stratified into Pre-LLM and Post-LLM eras. SciZoom provides three hierarchical summarization targets (Abstract, Contributions, and TL;DR) achieving compression ratios up to 600:1, enabling both multi-granularity summarization research and temporal mining of scientific writing patterns. Our linguistic analysis reveals striking shifts in phrase patterns (up to 10x for formulaic expressions) and rhetorical style (23% decline in hedging), suggesting that LLM-assisted writing produces more confident yet homogenized prose. SciZoom serves as both a challenging benchmark and a unique resource for mining the evolution of scientific discourse in the generative AI era. Our code and dataset are publicly available on GitHub (https://github.com/janghana/SciZoom) and Hugging Face (https://huggingface.co/datasets/hanjang/SciZoom), respectively.

Han Jang, Junhyeok Lee, Kyu Sung Choi

Seoul National University · Seoul National University Hospital · Seoul National University College of Medicine

LLM Evaluation Scientific Communication Text Summarization

Mar 17, 2026

Assessing the Pedagogical Readiness of Large Language Models as AI Tutors in Low-Resource Contexts: A Case Study of Nepal's K-10 Curriculum

The integration of Large Language Models (LLMs) into educational ecosystems promises to democratize access to personalized tutoring, yet the readiness of these systems for deployment in non-Western, low-resource contexts remains critically under-examined. This study presents a systematic evaluation of four state-of-the-art LLMs--GPT-4o, Claude Sonnet 4, Qwen3-235B, and Kimi K2--assessing their capacity to function as AI tutors within the specific curricular and cultural framework of Nepal's Grade 5-10 Science and Mathematics education. We introduce a novel, curriculum-aligned benchmark and a fine-grained evaluation framework inspired by the "natural language unit tests" paradigm, decomposing pedagogical efficacy into seven binary metrics: Prompt Alignment, Factual Correctness, Clarity, Contextual Relevance, Engagement, Harmful Content Avoidance, and Solution Accuracy. Our results reveal a stark "curriculum-alignment gap." While frontier models (GPT-4o, Claude Sonnet 4) achieve high aggregate reliability (approximately 97%), significant deficiencies persist in pedagogical clarity and cultural contextualization. We identify two pervasive failure modes: the "Expert's Curse," where models solve complex problems but fail to explain them clearly to novices, and the "Foundational Fallacy," where performance paradoxically degrades on simpler, lower-grade material due to an inability to adapt to younger learners' cognitive constraints. Furthermore, regional models like Kimi K2 exhibit a "Contextual Blindspot," failing to provide culturally relevant examples in over 20% of interactions. These findings suggest that off-the-shelf LLMs are not yet ready for autonomous deployment in Nepalese classrooms. We propose a "human-in-the-loop" deployment strategy and offer a methodological blueprint for curriculum-specific fine-tuning to align global AI capabilities with local educational needs.

Pratyush Acharya, Prasansha Bharati, Yokibha Chapagain +2

AI in Education LLM Evaluation Intelligent Tutoring Systems Cross-Cultural Language Model Evaluation

Mar 16, 2026

VIBEPASS: Can Vibe Coders Really Pass the Vibe Check?

As Large Language Models shift the programming toward human-guided ''vibe coding'', agentic coding tools increasingly rely on models to self-diagnose and repair their own subtle faults -- a capability central to autonomous software engineering yet never systematically evaluated. We present \name{}, the first empirical decomposition that jointly evaluates two coupled tasks: Fault-Triggering Test Generation (FT-Test) constructing a discriminative witness that exposes a latent bug, and Fault-targeted Program Repair (FPR), repairing it under varying diagnostic conditions. \name{} pairs competitive programming problems with LLM-generated solutions that pass partial test suites but fail on semantic edge cases, enabling controlled identification of where the diagnostic chain breaks down. Evaluating 12 frontier LLMs, we find that fault-targeted reasoning does not scale with general coding ability. Models produce syntactically valid test inputs at near-ceiling rates yet collapse on discriminative generation, with fault hypothesis generation -- not output validation -- as the dominant bottleneck. Test-guided repair reveals a complementary insight: when self-generated tests successfully witness a fault, the resulting repair matches or outperforms repair guided by externally provided tests, but tests that fail to witness the fault actively degrade repair below unguided baselines. Together, these results reframe the challenge of autonomous debugging: the binding bottleneck is not code synthesis or test validity but fault-target reasoning, a capability that remains deficient across all frontier models. As Large Language Models shift the programming toward human-guided ''vibe coding'', agentic coding tools increasingly rely on models to self-diagnose and repair their own subtle faults -- a capability central to autonomous software engineering yet never systematically evaluated.

Srijan Bansal, Jiao Fangkai, Yilun Zhou +3

Salesforce AI Research · Nanyang Technological University · A*STAR

LLM Self-Correction Language Model Error Detection LLM Evaluation Automated Program Repair Automated Test Generation

Mar 16, 2026

Temporal Fact Conflicts in LLMs: Reproducibility Insights from Unifying DYNAMICQA and MULAN

Large Language Models (LLMs) often struggle with temporal fact conflicts due to outdated or evolving information in their training data. Two recent studies with accompanying datasets report opposite conclusions on whether external context can effectively resolve such conflicts. DYNAMICQA evaluates how effective external context is in shifting the model's output distribution, finding that temporal facts are more resistant to change. In contrast, MULAN examines how often external context changes memorised facts, concluding that temporal facts are easier to update. In this reproducibility paper, we first reproduce experiments from both benchmarks. We then reproduce the experiments of each study on the dataset of the other to investigate the source of their disagreement. To enable direct comparison of findings, we standardise both datasets to align with the evaluation settings of each study. Importantly, using an LLM, we synthetically generate realistic natural language contexts to replace MULAN's programmatically constructed statements when reproducing the findings of DYNAMICQA. Our analysis reveals strong dataset dependence: MULAN's findings generalise under both methodological frameworks, whereas applying MULAN's evaluation to DYNAMICQA yields mixed outcomes. Finally, while the original studies only considered 7B LLMs, we reproduce these experiments across LLMs of varying sizes, revealing how model size influences the encoding and updating of temporal facts. Our results highlight how dataset design, evaluation metrics, and model size shape LLM behaviour in the presence of temporal knowledge conflicts.

Ritajit Dey, Iadh Ounis, Graham McDonald +1

University of Glasgow, Glasgow, UK · University of Strathclyde, Glasgow, UK

Knowledge Conflicts in Language Models Temporal Reasoning in Language Models Belief Updating in LLMs LLM Evaluation

Mar 16, 2026

When Stability Fails: Hidden Failure Modes Of LLMS in Data-Constrained Scientific Decision-Making

Large language models (LLMs) are increasingly used as decision-support tools in data-constrained scientific workflows, where correctness and validity are critical. However, evaluation practices often emphasize stability or reproducibility across repeated runs. While these properties are desirable, stability alone does not guar- antee agreement with statistical ground truth when such references are available. We introduce a controlled behavioral evaluation framework that explicitly sep- arates four dimensions of LLM decision-making: stability, correctness, prompt sensitivity, and output validity under fixed statistical inputs. We evaluate multi- ple LLMs using a statistical gene prioritization task derived from differential ex- pression analysis across prompt regimes involving strict and relaxed significance thresholds, borderline ranking scenarios, and minor wording variations. Our ex- periments show that LLMs can exhibit near-perfect run-to-run stability while sys- tematically diverging from statistical ground truth, over-selecting under relaxed thresholds, responding sharply to minor prompt wording changes, or producing syntactically plausible gene identifiers absent from the input table. Although sta- bility reflects robustness across repeated runs, it does not guarantee agreement with statistical ground truth in structured scientific decision tasks. These findings highlight the importance of explicit ground-truth validation and output validity checks when deploying LLMs in automated or semi-automated scientific work- flows.

Nazia Riasat

North Dakota State University

LLM Evaluation LLM Reliability Language Model Generation Evaluation Prompt Sensitivity

Mar 16, 2026

Establishing Construct Validity in LLM Capability Benchmarks Requires Nomological Networks

Recent work in machine learning increasingly attributes human-like capabilities such as reasoning or theory of mind to large language models (LLMs) on the basis of benchmark performance. This paper examines this practice through the lens of construct validity, understood as the problem of linking theoretical capabilities to their empirical measurements. It contrasts three influential frameworks: the nomological account developed by Cronbach and Meehl, the inferential account proposed by Messick and refined by Kane, and Borsboom's causal account. I argue that the nomological account provides the most suitable foundation for current LLM capability research. It avoids the strong ontological commitments of the causal account while offering a more substantive framework for articulating construct meaning than the inferential account. I explore the conceptual implications of adopting the nomological account for LLM research through a concrete case: the assessment of reasoning capabilities in LLMs.

Timo Freiesleben

Munich Center for Mathematical Philosophy (MCMP) LMU Munich

Mar 16, 2026

Pretraining and Benchmarking Modern Encoders for Latvian

Encoder-only transformers remain essential for practical NLP tasks. While recent advances in multilingual models have improved cross-lingual capabilities, low-resource languages such as Latvian remain underrepresented in pretraining corpora, and few monolingual Latvian encoders currently exist. We address this gap by pretraining a suite of Latvian-specific encoders based on RoBERTa, DeBERTaV3, and ModernBERT architectures, including long-context variants, and evaluating them across a diverse set of Latvian diagnostic and linguistic benchmarks. Our models are competitive with existing monolingual and multilingual encoders while benefiting from recent architectural and efficiency advances. Our best model, lv-deberta-base (111M parameters), achieves the strongest overall performance, outperforming larger multilingual baselines and prior Latvian-specific encoders. We release all pretrained models and evaluation resources to support further research and practical applications in Latvian NLP.

Arturs Znotins

Institute of Mathematics and Computer Science, University of Latvia, 29 Raina bulv., Riga, LV-1459, Latvia

Language Model Pretraining Low-Resource Language Processing Language Modeling LLM Evaluation

Mar 16, 2026

Distributional Open-Ended Evaluation of LLM Cultural Value Alignment Based on Value Codebook

As LLMs are globally deployed, aligning their cultural value orientations is critical for safety and user engagement. However, existing benchmarks face the Construct-Composition-Context (

C^3

) challenge: relying on discriminative, multiple-choice formats that probe value knowledge rather than true orientations, overlook subcultural heterogeneity, and mismatch with real-world open-ended generation. We introduce DOVE, a distributional evaluation framework that directly compares human-written text distributions with LLM-generated outputs. DOVE utilizes a rate-distortion variational optimization objective to construct a compact value codebook from 10K documents, mapping text into a structured value space to filter semantic noise. Alignment is measured using unbalanced optimal transport, capturing intra-cultural distributional structures and subgroup diversity. Experiments across 12 LLMs show that DOVE achieves superior predictive validity, attaining a 31.56% correlation with downstream tasks, while maintaining high reliability with as few as 500 samples per culture.

Jaehyeok Lee, Xiaoyuan Yi, Jing Yao +4

Sungkyunkwan University · Microsoft Research Asia · Singapore University of Technology and Design

LLM Evaluation LLM Alignment Cultural Alignment Cross-Cultural Language Model Evaluation

Mar 16, 2026

LLMs as Signal Detectors: Sensitivity, Bias, and the Temperature-Criterion Analogy

Large language models (LLMs) are evaluated for calibration using metrics such as Expected Calibration Error that conflate two distinct components: the model's ability to discriminate correct from incorrect answers (sensitivity) and its tendency toward confident or cautious responding (bias). Signal Detection Theory (SDT) decomposes these components. While SDT-derived metrics such as AUROC are increasingly used, the full parametric framework - unequal-variance model fitting, criterion estimation, z-ROC analysis - has not been applied to LLMs as signal detectors. In this pre-registered study, we treat three LLMs as observers performing factual discrimination across 168,000 trials and test whether temperature functions as a criterion shift analogous to payoff manipulations in human psychophysics. Critically, this analogy may break down because temperature changes the generated answer itself, not only the confidence assigned to it. Our results confirm the breakdown with temperature simultaneously increasing sensitivity (AUC) and shifting criterion. All models exhibited unequal-variance evidence distributions (z-ROC slopes 0.52-0.84), with instruct models showing more extreme asymmetry (0.52-0.63) than the base model (0.77-0.87) or human recognition memory (~0.80). The SDT decomposition revealed that models occupying distinct positions in sensitivity-bias space could not be distinguished by calibration metrics alone, demonstrating that the full parametric framework provides diagnostic information unavailable from existing metrics.

Jon-Paul Cacioli

Independent Researcher, Melbourne, Australia

LLM Evaluation Language Model Calibration Confidence Estimation in Language Models

Mar 16, 2026

BrainBench: Exposing the Commonsense Reasoning Gap in Large Language Models

Large language models (LLMs) achieve impressive scores on standard benchmarks yet routinely fail questions that any human would answer correctly in seconds. We introduce BrainBench, a benchmark of 100 brainteaser questions spanning 20 carefully designed categories, each targeting a specific commonsense reasoning failure mode in LLMs. Categories range from implicit physical constraints ("Should I walk or drive my rental car to the return lot?") to semantic scope tricks and default assumption hijacks. We evaluate eight frontier models -- four from the Claude family and four from the GPT family -- using a zero-shot protocol with 10 independent runs per question. The best model, Claude Opus 4.6 with extended thinking, achieves only 80.3% accuracy; the worst, GPT-4o, scores 39.7%. Even top-performing models exhibit a 6-16 percentage-point gap between accuracy and consistency, revealing stochastic reasoning. Cross-lingual evaluation in Chinese shows most models degrade by 2-8 percentage points, confirming that these failures reflect reasoning deficits rather than language-specific artifacts. BrainBench provides a fine-grained diagnostic tool for identifying where and why LLMs substitute surface heuristics for genuine commonsense reasoning.

Yuzhe Tang

Georgia Institute of Technology

Commonsense Reasoning Language Model Reasoning Evaluation LLM Evaluation Implicit Reasoning in Language Models Reasoning Consistency in Language Models

Mar 15, 2026

CangjieBench: Benchmarking LLMs on a Low-Resource General-Purpose Programming Language

Large Language Models excel in high-resource programming languages but struggle with low-resource ones. Existing research related to low-resource programming languages primarily focuses on Domain-Specific Languages (DSLs), leaving general-purpose languages that suffer from data scarcity underexplored. To address this gap, we introduce CangjieBench, a contamination-free benchmark for Cangjie, a representative low-resource general-purpose language. The benchmark comprises 248 high-quality samples manually translated from HumanEval and ClassEval, covering both Text-to-Code and Code-to-Code tasks. We conduct a systematic evaluation of diverse LLMs under four settings: Direct Generation, Syntax-Constrained Generation, Retrieval-Augmented Generation (RAG), and Agent. Experiments reveal that Direct Generation performs poorly, whereas Syntax-Constrained Generation offers the best trade-off between accuracy and computational cost. Agent achieve state-of-the-art accuracy but incur high token consumption. Furthermore, we observe that Code-to-Code translation often underperforms Text-to-Code generation, suggesting a negative transfer phenomenon where models overfit to the source language patterns. We hope that our work will offer valuable insights into LLM generalization to unseen and low-resource programming languages. Our code and data are available at https://github.com/cjhCoder7/CangjieBench.

Junhang Cheng, Fang Liu, Jia Li +3

Beihang University, Beijing, China · Wuhan University, Wuhan, China

Code Generation LLM Code Generation Low-Resource Language Processing LLM Evaluation Code Translation

Mar 15, 2026

An Industrial-Scale Insurance LLM Achieving Verifiable Domain Mastery and Hallucination Control without Competence Trade-offs

Adapting Large Language Models (LLMs) to high-stakes vertical domains like insurance presents a significant challenge: scenarios demand strict adherence to complex regulations and business logic with zero tolerance for hallucinations. Existing approaches often suffer from a Competency Trade-off - sacrificing general intelligence for domain expertise - or rely heavily on RAG without intrinsic reasoning. To bridge this gap, we present INS-S1, an insurance-specific LLM family trained via a novel end-to-end alignment paradigm. Our approach features two methodological innovations: (1) A Verifiable Data Synthesis System that constructs hierarchical datasets for actuarial reasoning and compliance; and (2) A Progressive SFT-RL Curriculum Framework that integrates dynamic data annealing with a synergistic mix of Verified Reasoning (RLVR) and AI Feedback (RLAIF). By optimizing data ratios and reward signals, this framework enforces domain constraints while preventing catastrophic forgetting. Additionally, we release INSEva, the most comprehensive insurance benchmark to date (39k+ samples). Extensive experiments show that INS-S1 achieves SOTA performance on domain tasks, significantly outperforming DeepSeek-R1 and Gemini-2.5-Pro. Crucially, it maintains top-tier general capabilities and achieves a record-low 0.6% hallucination rate (HHEM). Our results demonstrate that rigorous domain specialization can be achieved without compromising general intelligence.

Qian Zhu, Xinnan Guo, Jingjing Huo +5

Ant Group

Domain Adaptation for LLMs LLM Alignment LLM Hallucination Mitigation RL for Language Model Reasoning LLM Evaluation