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.

Jul 14, 2025

Detoxify: A framework for abusive text transformation using LLMs

Although Large Language Models (LLMs) have demonstrated significant advancements in natural language processing tasks, their effectiveness in the classification and transformation of abusive text into non-abusive versions remains an area for exploration. In this study, we present Detoxify: a framework that employs LLMs to transform abusive text (tweets and reviews) containing hate speech and profanity into non-abusive text while retaining the original intent. We evaluate the performance of four state-of-the-art LLMs, such as Gemini, GPT-4o, DeekSeek and Groq, on their ability to identify abusive text. We aim to transform and obtain a text that is clean of abusive and inappropriate content, but maintains a similar level of sentiment and semantics, i.e. the transformed text needs to maintain its message. Afterwards, we evaluate the raw and transformed datasets with sentiment analysis and semantic analysis. Our results show Groq provides vastly different results when compared with other LLMs. We have identified similarities between GPT-4o and DeepSeek. Groq stood out as the most distinct, as it often restructured sentences with excessive positive phrasing, with the original context lost or altered.

Rohitash Chandra, Jiyong Choi, Jayesh Sonawane

Transitional Artificial Intelligence Research Group, School of Mathematics and Statistics, UNSW, Sydney, Australia · Centre for Artificial Intelligence and Innovation, Pingla Institute, Sydney, Australia · Centre for Artificial Intelligence and Innovation, Pingla

LLM Evaluation Text Detoxification

Jul 12, 2025

LLM Bidders Preserve the Mechanism-Level Orderings of Human Bidders

Training on vast amounts of human-generated data has motivated growing interest in using large language models (LLMs) to simulate human behavior. We ask which features of human behavior general-purpose models preserve when used out of the box in auctions, where multiple bidders interact under explicit rules and incentives. We evaluate five LLMs across seven laboratory settings against human benchmarks reconstructed from published experiments, with uncertainty bands for the private-value comparisons. Our main focus is on three large models without extended test-time reasoning: GPT-4o, Claude~~3.5 Haiku, and Gemini~~2.0 Flash. LLM and human deviations from theory differ in magnitude and often in direction: humans overbid in second-price auctions, whereas most models that deviate underbid. Surprisingly, without task-specific fine-tuning or calibration to human bids, the three non-reasoning large models robustly preserve key orderings of auction formats by deviation from theory. First-price auctions are harder than second-price, and ascending clocks reduce deviations relative to sealed bids wherever data are adequate. Kendall's

τ_b

between the human and GPT-4o difficulty rankings is

0.60

and positive in every joint bootstrap draw. The reasoning model bids almost at equilibrium in the observed private-value settings, leaving little variation in errors to compare; the small model's large errors yield an inverted ranking. All five models nevertheless reproduce the stronger first-price winner's curse. Clock framing improves bidding for two of the three non-reasoning large models, and GPT-4o recovers the ordering of last-minute bidding across closing rules in an eBay-style marketplace.

Anand Shah, Kehang Zhu, Yanchen Jiang +4

MIT & NBER

LLM Evaluation Human Behavior Simulation LLM Decision-Making

Jun 24, 2025

LLM-Based Social Simulations Require a Boundary

This position paper argues that LLM-based social simulations require clear boundaries to make meaningful contributions to social science. While Large Language Models (LLMs) offer promising capabilities for simulating human behavior, their tendency to produce homogeneous outputs, acting as an "average persona", fundamentally limits their ability to capture the behavioral diversity essential for complex social dynamics. We examine why heterogeneity matters for social simulations and how current LLMs fall short, analyzing the relationship between mean alignment and variance in LLM-generated behaviors. Through a systematic review of representative studies, we find that validation practices often fail to match the heterogeneity requirements of research questions: while most papers include ground truth comparisons, fewer than half explicitly assess behavioral variance, and most that do report lower variance than human populations. We propose that researchers should: (1) match validation depth to the heterogeneity demands of their research questions, (2) explicitly report variance alongside mean alignment, and (3) constrain claims to collective-level qualitative patterns when variance is insufficient. Rather than dismissing LLM-based simulation, we advocate for a boundary-aware approach that ensures these methods contribute genuine insights to social science.

Zengqing Wu, Run Peng, Takayuki Ito +2

University of Osaka · University of Michigan · Kyoto University +1

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

Jun 19, 2025

Measuring Intent Comprehension in LLMs

People judge interactions with large language models (LLMs) as successful when outputs match what they want, not what they type. Yet LLMs are trained to predict the next token solely from text input, not underlying intent. Because written language is an imperfect proxy for intent, and correlations between phrasing and desired outcomes can break down in training data, models that rely too heavily on surface cues may respond inconsistently to semantically equivalent prompts. This makes it essential to evaluate whether LLMs can reliably infer user intent-especially in high-stakes settings where robustness and generalization are critical. We introduce a formal framework for assessing intent comprehension in LLMs: whether a model demonstrates robust understanding of user intent by producing consistent outputs across semantically equivalent prompts while differentiating between prompts with distinct intents. Our evaluation approach is based on a variance decomposition of model responses into three components: variability due to user intent, user articulation, and model uncertainty. Models that understand what users want, and are not overly sensitive to textual cues, should attribute most output variance to intent differences, rather than articulation style. Applying this framework across diverse domains, we find that, within the five LLaMA and Gemma models we evaluate, larger models typically assign a greater share of variance to intent, indicating stronger comprehension of intent, although gains are uneven and often modest with increasing model size. These results motivate moving beyond accuracy-only benchmarks toward semantic diagnostics that directly assess whether models understand what users intend.

Nadav Kunievsky, James A. Evans

Knowlesdge Lab, University of Chicago, Chicago, Illinois, USA · Knowledge Lab, Data Science Institute, Department of Sociology, University of Chicago, Chicago, Illinois, USA.

Prompt Sensitivity LLM Evaluation Intent Classification Language Model Robustness

Jun 4, 2025

LLMs are Brittle to Simple Code Transformations: Introducing CETBench - A Benchmark for Code-Equivalence Checking

We study how well LLMs can determine whether two programs are functionally equivalent. This is an important problem because benchmarking code equivalence helps assess LLM capability in tasks such as code rewriting and translation. To this end, we introduce CETBench - Code Equivalence with Transformations Benchmark - built from a repository of programs that may solve the same or different tasks. Each dataset instance is created by sampling a program pair and applying a random sequence of predefined code transformations, yielding either equivalent or non-equivalent pairs. Our analysis shows that even simple transformations cause a significant performance drop in state-of-the-art LLMs on code-equivalence checking. These challenges are further amplified in the cross-lingual setting when comparing programs written in different languages. To remedy this, we present a simple fine-tuning-based approach to boost LLM performance on the transformed pairs of programs. Our approach for dataset generation is generic, supporting cross-lingual equivalence checking, the generation of program pairs with varying difficulty levels, and the application of diverse transformations. In our experiments, we perform ablations over the difficulty level of original programs, as well as the kind of transformations used in generating pairs for equivalence checking. Our analysis presents deep insights into the working of LLMs for the task of code-equivalence, and points to the fact that they may still be far from what could be termed as a semantic understanding of the underlying code.

Neeva Oza, Ishaan Govil, Parul Gupta +3

Indian Institute of Technology Delhi · IBM Research AI

LLM Evaluation Language Model Robustness Code Language Models

Jun 2, 2025

C-VARC: A Large-Scale Chinese Value Rule Corpus for Value Alignment of Large Language Models

Aligning LLMs with Chinese values requires culturally grounded resources that connect abstract values to concrete behavioral guidance. We introduce C-VARC, a Chinese Value Rule Corpus containing 257,609 Rules of Thumb (RoTs) organized into three dimensions, 12 core values, and 50 derived values. Among them, 36,000 rules underwent independent human review by two annotators (Cohen's kappa>0.85), with disagreements adjudicated by a third. C-VARC supports reusable rule-driven pipelines for both value evaluation and model alignment. Compared with abstract-value guidance, RoT guidance produces greater semantic and character-level n-gram diversity. Across five non-political sensitive themes, the seven-model average selection rate for C-VARC-consistent options is at least 72.0%, while five Chinese annotators show overall rates of at least 87.5%. Conflicting RoT pairs further enable the construction of 404,505 moral dilemmas, with 7,191 of 10,998 evaluated cases eliciting cross-model disagreement. For alignment, C-VARC-Query consistently outperforms matched Value-only supervision on CMoralEval across both Qwen3-8B and Llama-3.1-8B-Instruct, with average gains of 2.97 and 12.82 percentage points, respectively, and positive paired gains across all three seeds for both backbones. It also improves CValues and C-Eval over Value-only supervision on both backbones. C-VARC thus provides a scalable Chinese normative resource for constructing both value-evaluation and alignment supervision.

Ping Wu, Guobin Shen, Dongcheng Zhao +6

BrainCog Lab, Institute of Automation, Chinese Academy of Sciences, Beijing, China · The School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China · Beijing Key Laboratory of Safe AI and Superalignment, Beijing, China +2

Value Alignment LLM Alignment Moral Reasoning in Language Models LLM Evaluation Cultural Knowledge in Language Models

May 31, 2025

Sarc7: Evaluating Sarcasm Detection and Generation with Seven Types and Emotion-Informed Techniques

Sarcasm is a form of humor where expressions convey meanings opposite to their literal interpretations. Classifying and generating sarcasm using large language models is vital for interpreting human communication. Sarcasm poses challenges for computational models, due to its nuanced nature. We introduce Sarc7, a benchmark that classifies 7 types of sarcasm: self-deprecating, brooding, deadpan, polite, obnoxious, raging, and manic by annotating entries of the MUStARD dataset. Classification was evaluated using zero-shot, few-shot, chain-of-thought (CoT), and a novel emotion-based prompting technique. We propose an emotion-based generation method developed by identifying key components of sarcasm-incongruity, shock value, and context dependency. Our classification experiments show that Gemini 2.5, using emotion-based prompting, outperforms other setups with an F1 score of 0.3664. Human evaluators preferred our emotion-based prompting, with 38.46% more successful generations than zero-shot prompting.

Raina Gao, Alyssa Jeong, Lang Xiong +4

Algoverse AI Research

LLM Evaluation Sarcasm Detection Text Generation

May 31, 2025

Benchmarks Are Not Atomic: Composition-Aware LLM Evaluation using BenchHub

LLM benchmarks are often treated as coherent measurement units, yet they are heterogeneous collections of instances spanning diverse domains, skills, formats, and contexts. As a result, aggregate benchmark scores can conflate model capability with benchmark composition, obscuring what existing benchmarks actually cover. We introduce BenchHub, a composition-aware evaluation framework that represents benchmarks as distributions over instance-level attributes. BenchHub integrates 54 benchmarks comprising 839K samples across 10 languages. It enables researchers to inspect benchmark contents, uncover reusable coverage hidden behind benchmark names, transparently compare benchmarks, and construct controllable evaluation sets for application-aligned model selection. Using BenchHub, we show that similarly motivated benchmarks can differ substantially in internal composition, existing benchmarks often contain reusable coverage beyond their stated purposes, models exhibit fine-grained category-level performance variation hidden by aggregate scores, and model rankings can shift under different reweighting and resampling configurations. Our results motivate evaluation practices that make benchmark composition explicit, inspectable, and controllable.

Eunsu Kim, Haneul Yoo, Guijin Son +3

KAIST · Yonsei University · OnelineAI +1

May 29, 2025

SocialMaze: A Benchmark for Evaluating and Enhancing Social Reasoning in Large Language Models in Complex Social Environments

Large language models (LLMs) are increasingly deployed in socially grounded applications, where success requires interpreting context, inferring others' mental states, and reasoning about unreliable information. Yet existing benchmarks rarely evaluate these demands jointly in complex, evolving settings. We introduce SocialMaze, a benchmark that organizes six tasks across social deduction games, daily-life interactions, and digital community platforms along three descriptive design axes: deep reasoning, dynamic interaction, and information uncertainty. These axes characterize intended sources of task difficulty rather than latent, factor-analytic dimensions of model capability. Automated checks and human validation support data quality. Evaluations of twelve proprietary and open-weight LLMs show substantial variation in the use of evolving interaction histories; stronger chain-of-thought reasoners perform better on tasks requiring deeper inference, while uncertainty consistently degrades performance. Reasoning workflows help weaker short-chain-of-thought backbones but saturate on stronger reasoners. Finally, targeted fine-tuning on curated reasoning traces substantially improves structured social-reasoning tasks, whereas transfer to language-aggregation tasks remains statistically inconclusive. The project homepage is available at https://xzx34.github.io/socialmaze/.

Zixiang Xu, Yanbo Wang, Yue Huang +13

Mohamed bin Zayed University of Artificial Intelligence · University of Southern California · University of Notre Dame +3

LLM Evaluation AI Agent Benchmarks Social Reasoning LLM Reasoning

May 29, 2025

PBEBench: A Multi-Step Programming by Examples Reasoning Benchmark inspired by Historical Linguistics

Although many benchmarks evaluate the reasoning abilities of Large Language Models (LLMs) within domains such as mathematics, coding, or data wrangling, few abstract away from domain specifics to examine reasoning as a capability in and of itself. We contribute a novel type of benchmark evaluating the inductive reasoning capabilities of LLMs that is inspired by the forward reconstruction task from historical linguistics but is formulated in an extremely simple, general way (in the form of Programming by Examples). The task involves generating a cascade of simple string rewrite programs to transform a given list of input strings into a list of desired output strings. We present a fully automated pipeline that programmatically generates problems of this type with controllable difficulty, enabling scalable evaluation of reasoning models while avoiding contamination. Using this approach, we construct two benchmarks: PBEBench-Lite, which efficiently stratifies models of varying capabilities, and PBEBench, which requires models to induce programs similar in complexity to those constructed by historical linguists. Our experiments reveal a substantial performance gap between models that leverage test-time compute or LCoT (long chain-of-thought) reasoning and those that do not. Moreover, although recent models show promise, the solve rate for both of them drops below 5% for hard instances of the PBEBench dataset (ground truth cascade lengths of 20 and 30, respectively), falling well short of realistic historical linguistics requirements even with computationally expensive, popular scaling techniques from the PBE and reasoning literature. Additionally, we also study the effectiveness of different scaling strategies and the impact of various hyperparameters on the difficulty of the generated data using gpt-oss-120b, the best-performing open-source model.

Atharva Naik, Prakam, Yash Mathur +6

Carnegie Mellon University · Independent Researcher · Ohio State University

Program Synthesis Inductive Inference LLM Evaluation Logical Reasoning Synthetic Benchmark Generation LLM Reasoning+1

May 27, 2025

Assessing and Refining ChatGPT's Performance in Identifying Targeting and Inappropriate Language: A Comparative Study

This study evaluates the effectiveness of ChatGPT, an advanced AI model for natural language processing, in identifying targeting and inappropriate language in online comments. With the increasing challenge of moderating vast volumes of user-generated content on social network sites, the role of AI in content moderation has gained prominence. We compared ChatGPT's performance against crowd-sourced annotations and expert evaluations to assess its accuracy, scope of detection, and consistency. Our findings show that ChatGPT performs well in detecting inappropriate content, and that almost the entire gain from iterative prompt refinement came from a single change --- simplifying the label scheme and moving to GPT-4 --- after which further adjustments to prompt structure and sampling temperature had little effect. Its performance on targeting language was weaker and less stable, with higher false positive rates than expert judgments before refinement and higher false negative rates after a change of model. This study contributes to the field by demonstrating the potential of AI models like ChatGPT to enhance automated content moderation systems while also identifying areas for further improvement. The results underscore the importance of careful prompt design and model selection, and of contextual understanding, in supporting automated moderation and mitigating harmful online behavior.

Baran Barbarestani, Isa Maks, Piek Vossen

Vrije Universiteit Amsterdam, Computational Linguistics and Text Mining Lab

Content Moderation Toxicity Detection LLM Evaluation Hate Speech Detection Prompt Engineering

May 21, 2025

UrduFactCheck: An Agentic Fact-Checking Framework for Urdu with Evidence Boosting and Benchmarking

The rapid adoption of Large Language Models (LLMs) has raised important concerns about the factual reliability of their outputs, particularly in low-resource languages such as Urdu. Existing automated fact-checking systems are predominantly developed for English, leaving a significant gap for the more than 200 million Urdu speakers worldwide. In this work, we present UrduFactBench and UrduFactQA, two novel hand-annotated benchmarks designed to enable fact-checking and factual consistency evaluation in Urdu. While UrduFactBench focuses on claim verification, UrduFactQA targets the factuality of LLMs in question answering. These resources, the first of their kind for Urdu, were developed through a multi-stage annotation process involving native Urdu speakers. To complement these benchmarks, we introduce UrduFactCheck, a modular fact-checking framework that incorporates both monolingual and translation-based evidence retrieval strategies to mitigate the scarcity of high-quality Urdu evidence. Leveraging these resources, we conduct an extensive evaluation of twelve LLMs and demonstrate that translation-augmented pipelines consistently enhance performance compared to monolingual ones. Our findings reveal persistent challenges for open-source LLMs in Urdu and underscore the importance of developing targeted resources. All code and data are publicly available at https://github.com/mbzuai-nlp/UrduFactCheck.

Sarfraz Ahmad, Hasan Iqbal, Momina Ahsan +6

MBZUAI

LLM Evaluation Factual Consistency Evaluation Automated Fact-Checking Low-Resource Language Processing Evidence Retrieval

May 19, 2025

R3: Robust Rubric-Agnostic Reward Models

Reward models are essential for aligning language model outputs with human preferences, yet existing approaches often lack both controllability and interpretability. These models are typically optimized for narrow objectives, limiting their generalizability to broader downstream tasks. Moreover, their scalar outputs are difficult to interpret without contextual reasoning. To address these limitations, we introduce R3, a novel reward modeling framework that is rubric-agnostic, generalizable across evaluation dimensions, and provides interpretable, reasoned score assignments. R3 enables more transparent and flexible evaluation of language models, supporting robust alignment with diverse human values and use cases. Our models, data, and code are available as open source at https://github.com/rubricreward/r3.

David Anugraha, Zilu Tang, Lester James V. Miranda +5

Department of Computer Science, Stanford University · Department of Computer Science, Boston University · Allen Institute for AI +4

Reward Modeling LLM Evaluation LLM Alignment Language Model Generation Evaluation

May 18, 2025

Enriching Patent Claim Generation with European Patent Dataset

Drafting patent claims is time-intensive, costly, and requires professional skill. Therefore, researchers have investigated large language models (LLMs) to assist inventors in writing claims. However, existing work has largely relied on datasets from the United States Patent and Trademark Office (USPTO). To enlarge research scope regarding various jurisdictions, drafting conventions, and legal standards, we introduce EPD, a European patent dataset. EPD presents rich textual data and structured metadata to support multiple patent-related tasks, including claim generation. This dataset enriches the field in three critical aspects: (1) Jurisdictional diversity: Patents from different offices vary in legal and drafting conventions. EPD fills a critical gap by providing a benchmark of European patents to enable more comprehensive evaluation. (2) Quality improvement: EPD offers high-quality granted patents with finalized and legally approved texts, whereas others consist of patent applications that are unexamined or provisional. Experiments show that LLMs fine-tuned on EPD significantly outperform those trained on previous datasets and even GPT-4o in claim quality and cross-domain generalization. (3) Real-world simulation: We propose a difficult subset of EPD to better reflect real-world challenges. Results reveal that all tested LLMs perform substantially worse on challenging samples, which highlights the need for future research.

Lekang Jiang, Chengzu Li, Stephan Goetz

University of Cambridge

Text Generation LLM Evaluation

May 11, 2025

Can LLM-based Financial Investing Strategies Outperform the Market in Long Run?

Large Language Models (LLMs) have recently been leveraged for asset pricing tasks and stock trading applications, enabling AI agents to generate investment decisions from unstructured financial data. However, most evaluations of LLM timing-based investing strategies are conducted on narrow timeframes and limited stock universes, overstating effectiveness due to survivorship and data-snooping biases. We critically assess their generalizability and robustness by proposing FINSABER, a backtesting framework evaluating timing-based strategies across longer periods and a larger universe of symbols. Systematic backtests over two decades and 100+ symbols reveal that previously reported LLM advantages deteriorate significantly under broader cross-section and over a longer-term evaluation. Our market regime analysis further demonstrates that LLM strategies are overly conservative in bull markets, underperforming passive benchmarks, and overly aggressive in bear markets, incurring heavy losses. These findings highlight the need to develop LLM strategies that are able to prioritise trend detection and regime-aware risk controls over mere scaling of framework complexity.

Weixian Waylon Li, Hyeonjun Kim, Mihai Cucuringu +1

AIAI, School of Informatics The University of Edinburgh Edinburgh, United Kingdom · Global Finance Research Center Sungkyunkwan University Seoul, Republic of Korea · Dept. of Mathematics; Dept. of Statistics & OMI University of California, Los Angeles; University of Oxford United States; United Kingdom

LLM Evaluation Quantitative Finance Algorithmic Trading

Apr 24, 2025

The Sparse Frontier: Sparse Attention Trade-offs in Transformer LLMs

Sparse attention offers a promising strategy to extend long-context capabilities in Transformer LLMs, yet its efficiency-accuracy trade-offs remain unclear due to the lack of comprehensive evaluation. We address this gap with the largest-scale empirical analysis to date of training-free sparse attention, evaluating six methods across multiple model families and sizes, sequences up to 128K tokens, and sparsity levels up to 0.95 (i.e.,

1/20

attention budget) on nine diverse tasks. We first organise the rapidly evolving landscape of sparse attention methods into a taxonomy along four design axes. Our analysis then yields actionable insights: 1) sparse attention is effective: larger sparse models outperform smaller dense ones at equivalent cost, improving the Pareto frontier; 2) for the training-free methods we study, fine-grained per-query importance estimation during prefilling remains impractical-due to both the cost of estimation and the lack of sparse kernels that translate fine-grained sparsity into wall-clock gains-forcing a task-dependent choice between global-to-token and block-to-block selection. Instead, during decoding, token-to-page selection becomes feasible, enabling better generalisation and higher sparsity tolerance; 3) longer sequences tolerate higher sparsity, suggesting that fixed-budget methods in production are suboptimal. Together, these findings provide practical guidance for deploying sparse attention and methodological recommendations for future evaluations. Our code is available at https://github.com/PiotrNawrot/sparse-frontier.

Piotr Nawrot, Robert Li, Renjie Huang +3

University of Edinburgh · Cohere · Meta

Transformer Inference LLM Evaluation Self-Attention Long-Context Language Modeling LLM Inference Acceleration Sparse Attention+1

Apr 21, 2025

Evaluating LLMs on Chinese Topic Constructions: A Research Proposal Inspired by Tian et al. (2024)

This paper proposes a framework for evaluating large language models (LLMs) on Chinese topic constructions, focusing on their sensitivity to island constraints. Drawing inspiration from Tian et al. (2024), we outline an experimental design for testing LLMs' grammatical knowledge of Mandarin syntax. While no experiments have been conducted yet, this proposal aims to provide a foundation for future studies and invites feedback on the methodology.

Xiaodong Yang

College of Foreign Languages Zhejiang University of Technology 288, Liuhe Rd., Xihu District, Hangzhou, Zhejiang, 310023 China

LLM Evaluation Language Model Probing

Apr 18, 2025

DMind Benchmark: Toward a Holistic Assessment of LLM Capabilities across the Web3 Domain

The Web3 ecosystem, underpinned by cryptographic primitives and decentralized consensus, represents a high-stakes environment where software vulnerabilities and incentive misalignments translate directly into financial loss. As Large Language Models (LLMs) are increasingly integrated into this domain for tasks ranging from smart contract auditing to decentralized finance analytics, ensuring their reliability is paramount. However, general-purpose benchmarks fail to capture the specialized reasoning required for these adversarial and protocol-driven settings. To bridge this gap, we introduce DMind Benchmark, a comprehensive evaluation suite designed to rigorously assess LLM proficiency across the Web3 stack. DMind Benchmark encompasses nine distinct subdomains (spanning infrastructure, smart contracts, token economics, etc.) and combines objective knowledge retrieval with complex open-ended reasoning tasks that emulate real-world operational challenges. We conduct an extensive evaluation of 31 leading proprietary and open-weights models, employing a contamination-aware pipeline and verifying the statistical robustness of our scoring protocol through rigorous cross-judge consistency checks. Our analysis reveals a critical dichotomy: while models demonstrate competence in foundational infrastructure concepts, they exhibit significant vulnerabilities in high-reasoning tasks such as security auditing. Furthermore, we provide a Pareto analysis to guide cost-effective deployment and demonstrate through adversarial experiments that high performance on DMind Benchmark necessitates genuine reasoning rather than superficial memorization. Since its open-source release in April 2025, DMind Benchmark achieved the #1 trending position on Hugging Face for nearly a week and accumulated over 13k downloads by June 2026, establishing itself as a standard for advancing secure and trustworthy AI in Web3.

Enhao Huang, Pengyu Sun, Shuxun Wang +13

Zhejiang University Hangzhou, China · The State Key Laboratory of Blockchain and Data Security · DMind.ai Singapore +1

LLM Evaluation Software Vulnerability Detection LLMs for Cybersecurity

Apr 10, 2025

SAGE: A Search-AuGmented Evaluation of Large Language Models on Free-Form QA

As Large Language Models (LLMs) become increasingly used for question-answering (QA), relying on static, pre-annotated references for evaluation poses significant challenges in cost, scalability, and completeness. Meanwhile, using LLMs themselves as evaluators without external grounding remains unreliable for objective tasks, as they systematically over-accept incorrect answers, fabricate supporting rationales, and degrade sharply on questions that fall outside their training data. We propose Search-AuGmented Evaluation (SAGE), a framework to assess LLM outputs without fixed ground-truth answers. Unlike conventional metrics that compare to static references or depend solely on LLM-as-a-judge knowledge, SAGE acts as an agent that actively retrieves and synthesizes external evidence. It iteratively generates web queries, collects information, summarizes findings, and refines subsequent searches through reflection. By reducing dependence on static reference-driven evaluation protocols, SAGE offers a scalable and adaptive alternative for evaluating the factuality of LLMs. Experimental results on multiple free-form QA benchmarks show that SAGE achieves substantial to perfect agreement with human evaluations.

Sher Badshah, Ali Emami, Hassan Sajjad

Dalhousie University · Emory University

LLM Evaluation Language Model Generation Evaluation Factual Consistency Evaluation Automated Evaluation

Mar 26, 2025

Physics: Benchmarking Foundation Models on University-Level Physics Problem Solving

We introduce Physics, a benchmark for evaluating foundation models on advanced physics problem solving. It contains 1034 annotated, open-ended problems from PhD-qualifying exams across six core physics domains. Solving these problems requires applying domain-specific knowledge to physical setups and carrying out multi-step mathematical reasoning. Each problem includes a worked solution and independently checked reference answers. An LLM-based grading protocol evaluates individual answers and complete multi-part problems, supporting numeric, symbolic, and text responses. We evaluate 53 proprietary and open-weight models. On test problems without figures, high accuracy is concentrated in a small group of thinking models, with the strongest open-weight models approaching proprietary performance. Most open-weight models still solve fewer than half of these problems. Our analysis of Qwen3.5-9B reveals failures in applying physics knowledge and interpreting figures, with misread figures affecting several subquestions.

Kaiyue Feng, Yilun Zhao, Yixin Liu +4

New York University · Yale University · Notre Dame University

LLM Evaluation Physical Reasoning Scientific Reasoning in Language Models Scientific Reasoning

Mar 25, 2025

Scaling Evaluation-time Compute with Reasoning Models as Evaluators

As language model (LM) outputs get more and more natural, it is becoming more difficult than ever to evaluate their quality. Simultaneously, increasing LMs' "thinking" time through scaling test-time compute has proven an effective technique to solve challenging problems in domains such as math and code. This raises a natural question: can an LM's evaluation capability also be improved by spending more test-time compute? To answer this, we investigate employing reasoning models-LMs that natively generate long chain-of-thought reasoning-as evaluators. Specifically, we examine methods to leverage more test-time compute by (1) using reasoning models, and (2) prompting these models to evaluate not only the response as a whole (i.e., outcome evaluation) but also assess each step in the response separately (i.e., process evaluation). In experiments, we observe that the evaluator's performance improves monotonically when generating more reasoning tokens, similar to the trends observed in LM-based generation. Furthermore, we use these more accurate evaluators to rerank multiple generations, and demonstrate that spending more compute at evaluation time can be as effective as using more compute at generation time in improving an LM's problem-solving capability.

Seungone Kim, Ian Wu, Jinu Lee +8

CMU · UIUC2 · KAIST AI3 +3

LLM Evaluation Language Model Generation Evaluation Inference-Time Scaling Large Reasoning Models Language Model Reasoning Evaluation

Mar 7, 2025

No Free Labels: Limitations of LLM-as-a-Judge Without Human Grounding

Reliable evaluation of large language models (LLMs) is critical as their deployment rapidly expands, particularly in high-stakes domains such as business and finance. The LLM-as-a-Judge framework, which uses prompted LLMs to evaluate response quality, is appealing due to its scalability, low cost, and strong correlations with human stylistic preferences. However, it remains unclear how accurately these methods can assess response quality in domains where correctness matters more than style. To address this gap, we introduce the Business and Finance Fundamentals Benchmark (BFF-Bench), a dataset of 160 challenging questions and long-form responses authored by financial professionals. These experts subsequently evaluated the correctness of 1,200 responses generated by a diverse set of LLMs on both BFF-Bench and a challenging subset of MT-Bench. With this expert-annotated dataset of judgments (VERDICTS), we analyze the agreement between a suite of automated grading methods and human experts. While we observe that LLM Judges are more reliable than other grading methods, our findings reveal a clear pattern in LLM Judge performance: when not provided with a correct reference, judges show high agreement with human experts only on questions the judges were able to correctly answer themselves. We demonstrate that providing the judges with expert-written references largely mitigates this issue, highlighting the limits of using LLM-as-a-Judge without any form of human verification.

Michael Krumdick, Charles Lovering, Varshini Reddy +2

Kensho Technologies · MIT

LLM Evaluation LLM-as-a-Judge Human-in-the-Loop Evaluation Finance

Feb 28, 2025

AnnoCaseLaw: Legal Annotation Benchmark of Civil Negligence

Language models offer a way to scale costly, labour-intensive expert annotation, but their ability to reproduce expert judgements remains uncertain. This is particularly challenging in law, where annotators must distinguish the reasoning supporting judicial decisions from other legally relevant discussion. We present AnnoCaseLaw as a densely annotated benchmark comprising

471

U.S. civil-negligence court opinions with more than

12,000

expert highlights across five annotation categories. We evaluate proprietary frontier, open-weight, legal-specialised, and decision models on their ability to reproduce these annotations. Models perform better on Facts and Outcome than on Application of Law to Facts, Procedural History, and Relevant Precedents. On Application of Law to Facts, proprietary frontier models achieve the strongest balance of reasoning coverage and precision, while the open-weight model Gemma 4 achieves comparable Macro-F1 to proprietary frontier models at a fraction of the estimated inference cost. Our findings highlight the importance of evaluating annotation coverage and cost alongside aggregate scores when assessing the feasibility of automated legal annotation.

Rilton Franzone, Magnus Sesodia, Alina Petrova +7

University of Oxford · Thomson Reuters Labs · Vienna University of Technology +1

Legal NLP LLM-Assisted Annotation LLM Evaluation Legal Reasoning

Jan 30, 2025

Large Language Models for Cryptocurrency Transaction Analysis: A Bitcoin Case Study

Cryptocurrencies are widely used, yet current methods for analyzing transactions often rely on opaque, black-box models. While these models may achieve high performance, their outputs are usually difficult to interpret and adapt, making it challenging to capture nuanced behavioral patterns. Large language models (LLMs) have the potential to address these gaps, but their capabilities in this area remain largely unexplored, particularly in cybercrime detection. In this paper, we test this hypothesis by applying LLMs to real-world cryptocurrency transaction graphs, with a focus on Bitcoin, one of the most studied and widely adopted blockchain networks. We introduce a three-tiered framework to assess LLM capabilities: foundational metrics, characteristic overview, and contextual interpretation. This includes a new, human-readable graph representation format, LLM4TG, and a connectivity-enhanced transaction graph sampling algorithm, CETraS. Together, they significantly reduce token requirements, transforming the analysis of multiple moderately large-scale transaction graphs with LLMs from nearly impossible to feasible under strict token limits. Experimental results demonstrate that LLMs have outstanding performance on foundational metrics and characteristic overview, where the accuracy of recognizing most basic information at the node level exceeds 98.50% and the proportion of obtaining meaningful characteristics reaches 95.00%. Regarding contextual interpretation, LLMs also demonstrate strong performance in classification tasks, even with very limited labeled data, where top-3 accuracy reaches 72.43% with explanations. While the explanations are not always fully accurate, they highlight the strong potential of LLMs in this domain. At the same time, several limitations persist, which we discuss along with directions for future research.

Yuchen Lei, Yuexin Xiang, Rafael Dowsley +5

School of Cyber Science and Engineering, Wuhan University, Wuhan 430072, China · Faculty of Information Technology, Monash University, Clayton, VIC 3800, Australia · CSIRO’s Data61, Eveleigh, NSW 2015, Australia +2

LLM Evaluation LLMs for Cybersecurity LLMs for Graph Learning

Jan 24, 2025

Humanity's Last Exam

Benchmarks are important tools for tracking the rapid advancements in large language model (LLM) capabilities. However, benchmarks are not keeping pace in difficulty: LLMs now achieve over 90% accuracy on popular benchmarks like MMLU, limiting informed measurement of state-of-the-art LLM capabilities. In response, we introduce Humanity's Last Exam (HLE), a multi-modal benchmark at the frontier of human knowledge, designed to be the final closed-ended academic benchmark of its kind with broad subject coverage. HLE consists of 2,500 questions across dozens of subjects, including mathematics, humanities, and the natural sciences. HLE is developed globally by subject-matter experts and consists of multiple-choice and short-answer questions suitable for automated grading. Each question has a known solution that is unambiguous and easily verifiable, but cannot be quickly answered via internet retrieval. State-of-the-art LLMs demonstrate low accuracy and calibration on HLE, highlighting a significant gap between current LLM capabilities and the expert human frontier on closed-ended academic questions. To inform research and policymaking upon a clear understanding of model capabilities, we publicly release HLE at https://lastexam.ai.

Long Phan, Alice Gatti, Ziwen Han +1178

Michael Pokorny · Quinn · Tony +1

LLM Evaluation Language Model Reasoning Evaluation

Nov 15, 2024

Compound-QA: A Benchmark for Evaluating LLMs on Compound Questions

Large language models (LLMs) demonstrate remarkable performance across various tasks, prompting researchers to develop diverse evaluation benchmarks. However, most benchmarks typically measure the ability of LLMs to respond to individual questions, neglecting the complex interactions in real-world applications. We introduce Compound Question Synthesis (CQ-Syn) to build Compound-QA, a benchmark targeting questions composed of multiple interrelated sub-questions. This benchmark is derived from existing QA datasets, annotated with proprietary LLMs, and verified by humans for accuracy. It encompasses five categories: Factual-Statement, Cause-and-Effect, Hypothetical-Analysis, Comparison-and-Selection, and Evaluation-and-Suggestion. It evaluates the LLM capability in terms of three dimensions, including understanding, reasoning, and knowledge. Evaluating nine open-source LLMs on Compound-QA reveals that their performance on compound questions is notably lower than on non-compound questions. We further explore strategies to enhance LLMs' handling of compound questions, and our results show that these methods substantially improve models' comprehension and reasoning abilities.

Yutao Hou, Yajing Luo, Zhiwen Ruan +4

Shanghai University of Finance and Economics · Southern University of Science and Technology · University of Edinburgh +1

Multi-Hop QA LLM Evaluation Synthetic Benchmark Generation

Nov 5, 2024

Growing a Tail: Increasing Output Diversity in Large Language Models

How diverse are the outputs of large language models when diversity is desired? We examine the diversity of responses of several language models to questions with multiple possible answers, comparing them with human responses. Our findings suggest that models' responses are highly concentrated, reflecting narrow, mainstream outputs, in comparison to humans, whose responses exhibit a much longer-tail. We examine three simple and practical ways to increase output diversity: 1) increasing generation randomness via temperature sampling; 2) prompting models to answer from diverse perspectives using a single prompt; 3) aggregating outputs from several models. We find that these interventions, especially when combined, can substantially increase output diversity, although single-model outputs generally remain less diverse than the human baseline. We discuss potential implications of these findings for future work in AI policy and governance that wishes to preserve cultural diversity, an essential building block of a democratic social fabric.

Michal Shur-Ofry, Bar Horowitz-Amsalem, Adir Rahamim +1

Law Faculty, Hebrew University of Jerusalem; Jerusalem, Israel. · Faculty of Computer Science, Technion – Israel Institute of Technology; Haifa, Israel.

LLM Evaluation LLM Prompting Output Diversity in Language Models

Sep 27, 2024

Evaluation of OpenAI o1: Opportunities and Challenges of AGI

This comprehensive study evaluates the performance of OpenAI's o1-preview large language model across a diverse array of complex reasoning tasks, spanning multiple domains, including computer science, mathematics, natural sciences, medicine, linguistics, and social sciences. Through rigorous testing, o1-preview demonstrated remarkable capabilities, often achieving human-level or superior performance in areas ranging from coding challenges to scientific reasoning and from language processing to creative problem-solving. Key findings include: -83.3% success rate in solving complex competitive programming problems, surpassing many human experts. -Superior ability in generating coherent and accurate radiology reports, outperforming other evaluated models. -100% accuracy in high school-level mathematical reasoning tasks, providing detailed step-by-step solutions. -Advanced natural language inference capabilities across general and specialized domains like medicine. -Impressive performance in chip design tasks, outperforming specialized models in areas such as EDA script generation and bug analysis. -Remarkable proficiency in anthropology and geology, demonstrating deep understanding and reasoning in these specialized fields. -Strong capabilities in quantitative investing. O1 has comprehensive financial knowledge and statistical modeling skills. -Effective performance in social media analysis, including sentiment analysis and emotion recognition. The model excelled particularly in tasks requiring intricate reasoning and knowledge integration across various fields. While some limitations were observed, including occasional errors on simpler problems and challenges with certain highly specialized concepts, the overall results indicate significant progress towards artificial general intelligence.

Tianyang Zhong, Zhengliang Liu, Yi Pan +74

LLM Evaluation Code Generation Mathematical Reasoning Large Reasoning Models

Sep 1, 2024

Polyrating: A Cost-Effective and Bias-Aware Rating System for LLM Evaluation

Rating-based human evaluation has become an essential tool to accurately evaluate the impressive performance of large language models (LLMs). However, current rating systems suffer from several important limitations: first, they fail to account for biases that significantly influence evaluation results, second, they require large and expensive preference datasets to obtain accurate ratings, and third, they do not facilitate meaningful comparisons of model ratings across different tasks. To address these issues, we introduce Polyrating, an expressive and flexible rating system based on maximum a posteriori estimation that enables a more nuanced and thorough analysis of model performance at lower costs. Polyrating can detect and quantify biases affecting human preferences, ensuring fairer model comparisons. Further, Polyrating can reduce the cost of human evaluations by up to

41\%

for new models and up to

77\%

for new tasks by leveraging existing benchmark scores. Lastly, Polyrating enables direct comparisons of ratings across different tasks, providing a comprehensive understanding of an LLMs' strengths, weaknesses, and relative performance across different applications.

Jasper Dekoninck, Maximilian Baader, Martin Vechev

Department of Computer Science, ETH Zurich, Switzerland

Jul 16, 2024

VLMEvalKit: An Open-Source Toolkit for Evaluating Large Multi-Modality Models

We present VLMEvalKit: an open-source toolkit for evaluating large multi-modality models based on PyTorch. The toolkit aims to provide a user-friendly and comprehensive framework for researchers and developers to evaluate existing multi-modality models and publish reproducible evaluation results. In VLMEvalKit, we implement over 450+ large multi-modality model configurations, including both proprietary APIs and open-source models, and support 330+ benchmarks across diverse multi-modal benchmarks. By implementing a single interface, new models can be easily added to the toolkit, while the toolkit automatically handles the remaining workloads, including data preparation, distributed inference, prediction post-processing, and metric calculation. VLMEvalKit has also evolved to a broader evaluation suite spanning video/audio, document understanding, GUI grounding, spatial reasoning, safety, scientific reasoning, and multi-turn dialogue. Based on the evaluation results obtained with the toolkit, we host the OpenVLM Leaderboard, a comprehensive leaderboard to track the progress of multi-modality learning research. The toolkit is released on https://github.com/open-compass/VLMEvalKit and is actively maintained.

Haodong Duan, Xinyu Fang, Junming Yang +41

VLM Evaluation LLM Evaluation Multimodal Foundation Models Multimodal Model Evaluation Multimodal Understanding MLLM Evaluation