LLM-as-a-Judge

Follow this topic Ask about this topic

LLM-as-a-Judge studies how language models can assess the quality, correctness, safety, or usefulness of other models’ outputs, from chatbot responses to agent trajectories and professional work. Common approaches use explicit rubrics or reference answers, pairwise and listwise preference rankings, multi-judge panels, and trained or calibrated reward models. These methods are tested against human ratings on benchmarks for instruction following, factuality, safety, and domain-specific tasks, with research also exploring uncertainty estimates, abstention, and targeted human review. A central challenge is that consistency or agreement among judges does not guarantee alignment with people: shared biases, prompt sensitivity, weak evidence grounding, and adversarial manipulation can make scores unreliable or misleading.

Latest papers

559 papers93 papers in the last four weeks, up 16% on the four weeks before. 0.5% of all new papers.

Mar 5, 2026

Towards Provably Unbiased LLM Judges via Bias-Bounded Evaluation

As AI models progress beyond simple chatbots into more complex workflows, we draw ever closer to the event horizon beyond which AI systems will be utilized in autonomous, self-maintaining feedback loops. Any autonomous AI system will depend on automated, verifiable rewards and feedback; in settings where ground truth is sparse or non-deterministic, one practical source of such rewards is an LLM-as-a-Judge. Although LLM judges continue to improve, the literature has yet to introduce systems capable of enforcing standards with strong guarantees, particularly when bias vectors are unknown or adversarially discovered. To remedy this issue, we propose average bias-boundedness (A-BB), an algorithmic framework which formally guarantees reductions of harm/impact as a result of any measurable bias in an LLM judge. Evaluating on Arena-Hard-Auto with four LLM judges, we achieve (tau=0.5, delta=0.01) bias-bounded guarantees while retaining 61-99% correlation with original rankings across formatting and schematic bias settings, with most judge-bias combinations exceeding 80%. The code to reproduce our findings is available at https://github.com/penfever/bias-bounded-evaluation.

Benjamin Feuer, Lucas Rosenblatt, Oussama Elachqar

1Stanford University · 3New York University

Bias in LLM-as-a-Judge LLM-as-a-Judge

Mar 5, 2026

Judge Reliability Harness: Stress Testing the Reliability of LLM Judges

We present the Judge Reliability Harness, an open source library for constructing validation suites that test the reliability of LLM judges. As LLM based scoring is widely deployed in AI benchmarks, more tooling is needed to efficiently assess the reliability of these methods. Given a benchmark dataset and an LLM judge configuration, the harness generates reliability tests that evaluate both binary judgment accuracy and ordinal grading performance for free-response and agentic task formats. We evaluate four state-of-the-art judges across four benchmarks spanning safety, persuasion, misuse, and agentic behavior, and find meaningful variation in performance across models and perturbation types, highlighting opportunities to improve the robustness of LLM judges. No judge that we evaluated is uniformly reliable across benchmarks using our harness. For example, our preliminary experiments on judges revealed consistency issues as measured by accuracy in judging another LLM's ability to complete a task due to simple text formatting changes, paraphrasing, changes in verbosity, and flipping the ground truth label in LLM-produced responses. The code for this tool is available at: https://github.com/RANDCorporation/judge-reliability-harness

Sunishchal Dev, Andrew Sloan, Joshua Kavner +2

RAND Corporation, Santa Monica, CA, USA

LLM-as-a-Judge LLM Evaluation Language Model Robustness

Mar 5, 2026

C2-Faith: Benchmarking LLM Judges for Causal and Coverage Faithfulness in Chain-of-Thought Reasoning

Large language models (LLMs) are increasingly used as judges of chain-of-thought (CoT) reasoning, yet it remains unclear whether they can reliably assess process faithfulness rather than merely answer plausibility. We introduce C2-Faith, a benchmark built from PRM800K that explicitly decomposes faithfulness into two complementary dimensions: causality (whether each step logically follows from prior context) and coverage (whether essential intermediate inferences are present). Using controlled perturbations, we construct examples with known causal error positions by replacing a single step with a logically inconsistent variant, and with controlled coverage deletions at varying rates, enabling direct measurement against reference labels. We evaluate three frontier LLM judges across three tasks: binary causal detection, causal step localization, and coverage scoring. Our results reveal that judge reliability is highly task-dependent, with no single model dominating across settings. While models often detect that an error exists, they struggle to accurately localize it, indicating a substantial gap between detection and attribution. Moreover, all judges systematically overestimate reasoning completeness, assigning high coverage scores even when substantial portions of intermediate reasoning are missing. These findings expose fundamental limitations of LLM judges in process-level evaluation and highlight the need for more reliable and calibrated methods when using LLMs to assess reasoning quality.

Avni Mittal, Rauno Arike

CoT Faithfulness LLM-as-a-Judge Faithfulness of Language Model Explanations

Mar 5, 2026

IF-RewardBench: Benchmarking Judge Models for Instruction-Following Evaluation

Instruction-following is a foundational capability of large language models (LLMs), with its improvement hinging on scalable and accurate feedback from judge models. However, the reliability of current judge models in instruction-following remains underexplored due to several deficiencies of existing meta-evaluation benchmarks, such as their insufficient data coverage and oversimplified pairwise evaluation paradigms that misalign with model optimization scenarios. To this end, we propose IF-RewardBench, a comprehensive meta-evaluation benchmark for instruction-following that covers diverse instruction and constraint types. For each instruction, we construct a preference graph containing all pairwise preferences among multiple responses based on instruction-following quality. This design enables a listwise evaluation paradigm that assesses the capabilities of judge models to rank multiple responses, which is essential in guiding model alignment. Extensive experiments on IF-RewardBench reveal significant deficiencies in current judge models and demonstrate that our benchmark achieves a stronger positive correlation with downstream task performance compared to existing benchmarks. Our codes and data are available at https://github.com/thu-coai/IF-RewardBench.

Bosi Wen, Yilin Niu, Cunxiang Wang +5

The Conversational Artificial Intelligence (CoAI) Group, Tsinghua University · Zhipu AI · University of Electronic Science and Technology of China

LLM-as-a-Judge Instruction Following Reward Model Evaluation Learning to Rank

Mar 4, 2026

Who Judges the Judge? Evaluating LLM-as-a-Judge for French Medical open-ended QA

Automatic evaluation of medical open-ended question answering (OEQA) remains challenging due to the need for expert annotations. We evaluate whether large language models (LLMs) can act as judges of semantic equivalence in French medical OEQA, comparing closed-access, general-purpose, and biomedical domain-adapted models. Our results show that LLM-based judgments are strongly influenced by the model that generated the answer, with agreement varying substantially across generators. Domain-adapted and large general-purpose models achieve the highest alignment with expert annotations. We further show that lightweight adaptation of a compact model using supervised fine-tuning (SFT) and Group Relative Policy Optimization (GRPO) substantially improves performance and reduces generator sensitivity, even with limited data. Overall, our findings highlight the need for generator-aware evaluation and suggest that carefully adapted small models can support scalable evaluation in low-resource medical settings.

Ikram Belmadani, Oumaima El Khettari, Pacôme Constant dit Beaufils +2

Aix-Marseille Univ., CNRS, LIS UMR 7020, 13000 Marseille, France · Nantes Univ., École Centrale Nantes, CNRS, LS2N, UMR 6004, 44000 Nantes, France · Nantes Université, CHU Nantes, PHU 11: Santé Publique, Clinique des données, INSERM, CIC 1413, 44000 Nantes, France +2

LLM-as-a-Judge Bias in LLM-as-a-Judge Clinical Language Model Evaluation Medical QA

Mar 2, 2026

LiveCultureBench: a Multi-Agent, Multi-Cultural Benchmark for Large Language Models in Dynamic Social Simulations

Large language models (LLMs) are increasingly deployed as autonomous agents, yet evaluations focus primarily on task success rather than cultural appropriateness or evaluator reliability. We introduce LiveCultureBench, a multi-cultural, dynamic benchmark that embeds LLMs as agents in a simulated town and evaluates them on both task completion and adherence to socio-cultural norms. The simulation models a small city as a location graph with synthetic residents having diverse demographic and cultural profiles. Each episode assigns one resident a daily goal while others provide social context. An LLM-based verifier generates structured judgments on norm violations and task progress, which we aggregate into metrics capturing task-norm trade-offs and verifier uncertainty. Using LiveCultureBench across models and cultural profiles, we study (i) cross-cultural robustness of LLM agents, (ii) how they balance effectiveness against norm sensitivity, and (iii) when LLM-as-a-judge evaluation is reliable for automated benchmarking versus when human oversight is needed.

Viet-Thanh Pham, Lizhen Qu, Thuy-Trang Vu +2

Department of Data Science & AI, Monash University

AI Agent Benchmarks Cross-Cultural Language Model Evaluation Normative Conformity in Language Models Large Language Model-Based Social Simulation LLM-as-a-Judge

Mar 2, 2026

CyclicJudge: Mitigating Judge Bias Efficiently in LLM-based Evaluation

LLM-as-judge evaluation has become standard practice for open-ended model assessment; however, judges exhibit systematic biases that cannot be averaged out by increasing the number of scenarios or generations. These biases are often similar in magnitude to the model differences that benchmarks are designed to detect, resulting in unreliable rankings when single-judge evaluations are used. We introduce a variance decomposition that partitions benchmark score variance into scenario, generation, judge, and residual components. Based on this analysis, CyclicJudge, a round-robin assignment of judges to scenarios, is demonstrated to be the optimal strategy for a fixed judge panel and judge-call budget: the score recovers the panel mean exactly while matching the cost of single-judge evaluation. Empirical results on MT-Bench and MindEval validate the effectiveness of CyclicJudge as predicted, across both general-purpose and domain-specific evaluation settings.

Ziyi Zhu, Olivier Tieleman, Alexey Bukhtiyarov +1

Slingshot AI · Department of Engineering, University of Cambridge

LLM Evaluation Bias in LLM-as-a-Judge LLM-as-a-Judge

Mar 2, 2026

RubricBench: Aligning Model-Generated Rubrics with Human Standards

As Large Language Model (LLM) alignment evolves from simple completions to complex, highly sophisticated generation, Reward Models are increasingly shifting toward rubric-guided evaluation to mitigate surface-level biases. However, the community lacks a unified benchmark to assess this evaluation paradigm, as existing benchmarks lack both the discriminative complexity and the ground-truth rubric annotations required for rigorous analysis. To bridge this gap, we introduce RubricBench, a curated benchmark with 1,147 pairwise comparisons specifically designed to assess the reliability of rubric-based evaluation. Our construction employs a multi-dimensional filtration pipeline to target hard samples featuring nuanced input complexity and misleading surface bias, augmenting each with expert-annotated, atomic rubrics derived strictly from instructions. Comprehensive experiments reveal a substantial capability gap between human-annotated and model-generated rubrics, indicating that even state-of-the-art models struggle to autonomously specify valid evaluation criteria, lagging considerably behind human-guided performance.

Qiyuan Zhang, Junyi Zhou, Yufei Wang +8

City University of Hong Kong · Tencent Hunyuan · McGill - Mila & Quebec AI Institute +2

LLM Evaluation Rubric-Based Evaluation Automatic Rubric Generation LLM Reliability LLM-as-a-Judge Rubric-Guided Generation

Mar 1, 2026

Subliminal Signals in Preference Labels

As AI systems approach superhuman capabilities, scalable oversight increasingly relies on LLM-as-a-judge frameworks where models evaluate and guide each other's training. A core assumption is that binary preference labels provide only semantic supervision about response quality. We challenge this assumption by demonstrating that preference labels can function as a covert communication channel. We show that even when a neutral student model generates semantically unbiased completions, a biased judge can transmit unintended behavioral traits through preference assignments, which even strengthen across iterative alignment rounds. Our findings suggest that robust oversight in superalignment settings requires mechanisms that can detect and mitigate subliminal preference transmission, particularly when judges may pursue unintended objectives.

Isotta Magistrali, Frédéric Berdoz, Sam Dauncey +1

ETH Zurich, Switzerland

LLM-as-a-Judge Preference Alignment Bias in LLM-as-a-Judge

Mar 1, 2026

CollabEval: Enhancing LLM-as-a-Judge via Multi-Agent Collaboration

Large Language Models (LLMs) have revolutionized AI-generated content evaluation, with the LLM-as-a-Judge paradigm becoming increasingly popular. However, current single-LLM evaluation approaches face significant challenges, including inconsistent judgments and inherent biases from pre-training data. To address these limitations, we propose CollabEval, a novel multi-agent evaluation framework that implements a three-phase Collaborative Evaluation process: initial evaluation, multi-round discussion, and final judgment. Unlike existing approaches that rely on competitive debate or single-model evaluation, CollabEval emphasizes collaboration among multiple agents with strategic consensus checking for efficiency. Our extensive experiments demonstrate that CollabEval consistently outperforms single-LLM approaches across multiple dimensions while maintaining robust performance even when individual models struggle. The framework provides comprehensive support for various evaluation criteria while ensuring efficiency through its collaborative design.

Yiyue Qian, Shinan Zhang, Yun Zhou +3

Amazon AWS Generative AI Innovation Center · Amazon AWS Bedrock

LLM Evaluation LLM-as-a-Judge Language Model Generation Evaluation Multi-LLM Collaboration Bias in LLM-as-a-Judge

Feb 28, 2026

Advancing Multimodal Judge Models through a Capability-Oriented Benchmark and MCTS-Driven Data Generation

Using Multimodal Large Language Models (MLLMs) as judges to achieve precise and consistent evaluations has gradually become an emerging paradigm across various domains. Evaluating the capability and reliability of MLLM-as-a-judge systems is therefore essential for ensuring trustworthy assessment. Existing judge benchmarks categorize samples by task types but fail to capture the fundamental judgment capabilities required for reliable evaluation. In this work, we introduce M-JudgeBench, a ten-dimensional capability-oriented benchmark designed to comprehensively assess the judgment abilities of MLLMs. Our benchmark decomposes evaluation into pairwise Chain-of-Thought (CoT) comparison, length bias avoidance, and process error detection tasks, jointly covering ten fine-grained subtasks. This design enables diagnosis of model reliability across reasoning styles, response lengths, and cross-model variations. Systematic evaluation uncovers the systematic weaknesses in existing MLLM-as-a-judge systems. To address this issue, we further propose Judge-MCTS, a data construction framework generating pairwise reasoning trajectories with various correctness and length. Using Judge-MCTS, we construct an MCTS-augmented dataset and train M-Judger, a series of strong judge models. Extensive experiments demonstrate the superiority of M-Judger on existing judge benchmarks as well as M-JudgeBench. Overall, our work establishes a more principled foundation for evaluating MLLM-as-a-judge through M-JudgeBench and Judge-MCTS framework, paving the way for future research on judge model evaluation and capability-driven judge training.

Zeyu Chen, Huanjin Yao, Ziwang Zhao +1

LLM Evaluation LLM-as-a-Judge Synthetic Data Generation Multimodal Large Language Models Multimodal Model Evaluation

Feb 28, 2026

Are LLMs Reliable Code Reviewers? Systematic Overcorrection in Requirement Conformance Judgement

Large language models (LLMs) have become essential tools in software development, widely used for requirements engineering, code generation and review tasks. Software engineers often rely on LLMs to verify if code implementation satisfy task requirements, thereby ensuring code robustness and accuracy. However, it remains unclear whether LLMs can reliably determine code against the given task descriptions, which is usually in a form of natural language specifications. In this paper, we uncover a systematic failure of LLMs in matching code to natural language requirements. Specifically, with widely adopted benchmarks and unified prompts design, we demonstrate that LLMs frequently misclassify correct code implementation as non-compliant or defective. Surprisingly, we find that more detailed prompt design, particularly with those requiring explanations and proposed corrections, leads to higher misjudgment rates, highlighting critical reliability issues for LLM-based code assistants. We further analyze the mechanisms driving these failures and evaluate the reliability of rationale-required judgments. Building on these findings, we propose a Fix-guided Verification Filter that treats the model proposed fix as executable counterfactual evidence, and validates the original and revised implementations using benchmark tests and spec-constrained augmented tests. Our results expose previously under-explored limitations in LLM-based code review capabilities, and provide practical guidance for integrating LLM-based reviewers with safeguards in automated review and development pipelines.

Haolin Jin, Huaming Chen

University of Sydney, Sydney, NSW, Australia.

Automated Code Review LLM Reliability LLM-as-a-Judge

Feb 28, 2026

Confusion-Aware Rubric Optimization for LLM-based Automated Grading

Accurate and unambiguous guidelines are critical for large language model (LLM) based graders, yet manually crafting these prompts is often sub-optimal as LLMs can misinterpret expert guidelines or lack necessary domain specificity. Consequently, the field has moved toward automated prompt optimization to refine grading guidelines without the burden of manual trial and error. However, existing frameworks typically aggregate independent and unstructured error samples into a single update step, resulting in "rule dilution" where conflicting constraints weaken the model's grading logic. To address these limitations, we introduce Confusion-Aware Rubric Optimization (CARO), a novel framework that enhances accuracy and computational efficiency by structurally separating error signals. CARO leverages the confusion matrix to decompose monolithic error signals into distinct modes, allowing for the diagnosis and repair of specific misclassification patterns individually. By synthesizing targeted "fixing patches" for dominant error modes and employing a diversity-aware selection mechanism, the framework prevents guidance conflict and eliminates the need for resource-heavy nested refinement loops. Empirical evaluations on teacher education and STEM datasets demonstrate that CARO significantly outperforms existing SOTA methods. These results suggest that replacing mixed-error aggregation with surgical, mode-specific repair yields robust improvements in automated assessment scalability and precision.

Yucheng Chu, Hang Li, Kaiqi Yang +4

Michigan State University · University of Southern California

Automated Grading Prompt Optimization Rubric-Based Evaluation LLM-as-a-Judge

Feb 27, 2026

When Metrics Disagree: Automatic Similarity vs. LLM-as-a-Judge for Clinical Dialogue Evaluation

As Large Language Models (LLMs) are increasingly integrated into healthcare to address complex inquiries, ensuring their reliability remains a critical challenge. Recent studies have highlighted that generic LLMs often struggle in clinical contexts, occasionally producing misleading guidance. To mitigate these risks, this research focuses on the domain-specific adaptation of Llama-2-7B using the Low-Rank Adaptation (LoRA) technique. By injecting trainable low-rank matrices into the Transformer layers, we efficiently adapted the model using authentic patient-physician transcripts while preserving the foundational knowledge of the base model. Our objective was to enhance precision and contextual relevance in responding to medical queries by capturing the specialized nuances of clinical discourse. Due to the resource-intensive nature of large-scale human validation, the model's performance was evaluated through a dual-track framework: Track A utilized traditional lexical similarity metrics (e.g., BLEU, ROUGE), while Track B employed an "LLM-as-a-Judge" paradigm using GPT-4 for semantic assessment. Our results demonstrate that while the LoRA-enhanced model achieved significant improvements across all quantitative lexical dimensions, a profound disagreement surfaced in the GPT-4 evaluation, which marginally favored the baseline model's conversational flow. This metric divergence underscores a pivotal finding: traditional automated scores may not fully reflect clinical utility. Consequently, we propose that while automated metrics and LLM judges serve as valuable developmental proxies, rigorous validation by human medical experts remains an indispensable requirement for the safe deployment of LLMs in healthcare settings.

Bian Sun, Zhenjian Wang, Orvill de la Torre +1

Heinz College, Carnegie Mellon University · School of Computer Science, Carnegie Mellon University

Clinical Language Model Evaluation Generative AI Evaluation Domain Adaptation for LLMs Medical QA LLM-as-a-Judge

Feb 27, 2026

LFQA-HP-1M: A Large-Scale Human Preference Dataset for Long-Form Question Answering

Long-form question answering (LFQA) demands nuanced evaluation of multi-sentence explanatory responses, yet existing metrics often fail to reflect human judgment. We present LFQA-HP-1M, a large-scale dataset comprising 1.3M human pairwise preference annotations for LFQA. We propose nine rubrics for answer quality evaluation, and show that simple linear models based on these features perform comparably to state-of-the-art LLM evaluators. We further examine transitivity consistency, positional bias, and verbosity biases in LLM evaluators and demonstrate their vulnerability to adversarial perturbations. Overall, this work provides one of the largest public LFQA preference datasets and a rubric-driven framework for transparent and reliable evaluation.

Rafid Ishrak Jahan, Fahmid Shahriar Iqbal, Sagnik Ray Choudhury

University of North Texas, Denton, Texas, USA

LLM Evaluation LLM-as-a-Judge Bias in LLM-as-a-Judge

Feb 24, 2026

QEDBENCH: Quantifying the Alignment Gap in Automated Evaluation of University-Level Mathematical Proofs

As Large Language Models (LLMs) saturate elementary benchmarks, the research frontier has shifted from generation to the reliability of automated evaluation. We demonstrate that standard "LLM-as-a-Judge" protocols suffer from a systematic Alignment Gap when applied to upper-undergraduate to early graduate level mathematics. To quantify this, we introduce QEDBench, the first large-scale dual-rubric alignment benchmark to systematically measure alignment with human experts on university-level math proofs by contrasting course-specific rubrics against expert common knowledge criteria. By deploying a dual-evaluation matrix (7 judges x 5 solvers) against 1,000+ hours of human evaluation, we reveal that certain frontier evaluators like Claude Opus 4.5, DeepSeek-V3, Qwen 2.5 Max, and Llama 4 Maverick exhibit significant positive bias (up to +0.18, +0.20, +0.30, +0.36 mean score inflation, respectively). Furthermore, we uncover a critical reasoning gap in the discrete domain: while Gemini 3.0 Pro achieves state-of-the-art performance (0.91 average human evaluation score), other reasoning models like GPT-5 Pro and Claude Sonnet 4.5 see their performance significantly degrade in discrete domains. Specifically, their average human evaluation scores drop to 0.72 and 0.63 in Discrete Math, and to 0.74 and 0.50 in Graph Theory. In addition to these research results, we also release QEDBench as a public benchmark for evaluating and improving AI judges. Our benchmark is publicly published at https://github.com/qqliu/Yale-QEDBench.

Santiago Gonzalez, Alireza Amiri Bavandpour, Peter Ye +48

Mathematical Reasoning Benchmarks LLM Evaluation LLM-as-a-Judge Language Model Bias Evaluation Automated Grading

Feb 23, 2026

Case-Aware LLM-as-a-Judge Evaluation for Enterprise-Scale RAG Systems

Enterprise Retrieval-Augmented Generation (RAG) assistants operate in multi-turn, case-based workflows such as technical support and IT operations, where evaluation must reflect operational constraints, structured identifiers (e.g., error codes, versions), and resolution workflows. Existing RAG evaluation frameworks are primarily designed for benchmark-style or single-turn settings and often fail to capture enterprise-specific failure modes such as case misidentification, workflow misalignment, and partial resolution across turns. We present a case-aware LLM-as-a-Judge evaluation framework for enterprise multi-turn RAG systems. The framework evaluates each turn using eight operationally grounded metrics that separate retrieval quality, grounding fidelity, answer utility, precision integrity, and case/workflow alignment. A severity-aware scoring protocol reduces score inflation and improves diagnostic clarity across heterogeneous enterprise cases. The system uses deterministic prompting with strict JSON outputs, enabling scalable batch evaluation, regression testing, and production monitoring. Through a comparative study of two instruction-tuned models across short and long workflows, we show that generic proxy metrics provide ambiguous signals, while the proposed framework exposes enterprise-critical tradeoffs that are actionable for system improvement.

Mukul Chhabra, Luigi Medrano, Arush Verma

Dell Technologies, Austin, TX, USA

RAG Evaluation Multi-Turn Dialogue Evaluation LLM-as-a-Judge Information Retrieval

Feb 20, 2026

Luna-2: Scalable Single-Token Evaluation with Small Language Models

Real-time guardrails require evaluation that is accurate, cheap, and fast - yet today's default, LLM-as-a-judge (LLMAJ), is slow, expensive, and operationally non-deterministic due to multi-token generation. We present Luna-2, a novel architecture that leverages decoder-only small language models (SLMs) into a deterministic evaluation model to reliably compute complex task-specific LLMAJ metrics (e.g. toxicity, hallucination, tool selection quality, etc.) at an accuracy at par or higher than LLMAJ using frontier LLMs while drastically reducing the cost and latency of computation. Each metric is implemented as a lightweight LoRA/PEFT head on top of a shared SLM backbone, enabling hundreds of specialized metrics to run concurrently on a single GPU, deployable locally next to AI systems in a privacy-preserving and latency optimizing manner. Across content safety and hallucination benchmarks, Luna-2 matches the accuracy of state-of-the-art LLM-based evaluators while reducing inference cost by over 80x and latency by over 20x. In this paper, we outline the model architecture, training methodology and report real-world empirical results on accuracy, latency, and throughput results. In production, Luna-2 is protecting 100M+ AI sessions and processing over 100B tokens per month for our customers with eval cost savings of over $30M annually.

Vatsal Goel, Rishon Dsouza, Nikhil Ega +4

Galileo AI, San Francisco, USA

LLM Evaluation Language Model Safety Evaluation LLM-as-a-Judge Small Language Models

Feb 20, 2026

"How Do I ...?": Procedural Questions Predominate Student-LLM Chatbot Conversations

Providing scaffolding through educational chatbots built on Large Language Models (LLM) has potential risks and benefits that remain an open area of research. When students navigate impasses, they ask for help by formulating impasse-driven questions. Within interactions with LLM chatbots, such questions shape the user prompts and drive the pedagogical effectiveness of the chatbot's response. This paper focuses on such student questions from two datasets of distinct learning contexts: formative self-study, and summative assessed coursework. We analysed 6,113 messages from both learning contexts, using 11 different LLMs and three human raters to classify student questions using four existing schemas. On the feasibility of using LLMs as raters, results showed moderate-to-good inter-rater reliability, with higher consistency than human raters. The data showed that 'procedural' questions predominated in both learning contexts, but more so when students prepare for summative assessment. These results provide a basis on which to use LLMs for classification of student questions. However, we identify clear limitations in both the ability to classify with schemas and the value of doing so: schemas are limited and thus struggle to accommodate the semantic richness of composite prompts, offering only partial understanding the wider risks and benefits of chatbot integration. In the future, we recommend an analysis approach that captures the nuanced, multi-turn nature of conversation, for example, by applying methods from conversation analysis in discursive psychology.

Alexandra Neagu, Marcus Messer, Peter Johnson +1

Imperial College London, London, United Kingdom

Generative AI in Education Educational Dialogue Systems Conversational Agents Text Classification LLM-as-a-Judge LLM-Assisted Annotation

Feb 18, 2026

References Improve LLM Alignment in Non-Verifiable Domains

While Reinforcement Learning with Verifiable Rewards (RLVR) has shown strong effectiveness in reasoning tasks, it cannot be directly applied to non-verifiable domains lacking ground-truth verifiers, such as LLM alignment. In this work, we investigate whether reference-guided LLM-evaluators can bridge this gap by serving as soft "verifiers". First, we design evaluation protocols that enhance LLM-based evaluators for LLM alignment using reference outputs. Through comprehensive experiments, we show that a reference-guided approach substantially improves the accuracy of less capable LLM-judges using references from frontier models; stronger LLM-judges can also be enhanced by high-quality (i.e., human-written) references. Building on these improved judges, we demonstrate the utility of high-quality references in alignment tuning, where LLMs guided with references are used as judges to self-improve. We show that reference-guided self-improvement yields clear gains over both direct SFT on reference outputs and self-improvement with reference-free judges, achieving performance comparable to training with ArmoRM, a strong finetuned reward model. Specifically, our method achieves 73.1% and 58.7% on AlpacaEval and Arena-Hard with Llama-3-8B-Instruct, and 70.0% and 74.1% with Qwen2.5-7B, corresponding to average absolute gains of +20.2 / +17.1 points over SFT distillation and +5.3 / +3.6 points over reference-free self-improvement on AlpacaEval / Arena-Hard. These results highlight the potential of using reference-guided LLM-evaluators to enable effective LLM post-training in non-verifiable domains.

Kejian Shi, Yixin Liu, Peifeng Wang +3

Yale University · Meta · Scale AI +2

LLM Alignment LLM-as-a-Judge LLM Self-Refinement LLM Evaluation

Feb 18, 2026

Who can we trust? LLM-as-a-jury for Comparative Assessment

Large language models (LLMs) are increasingly applied as automatic evaluators for natural language generation assessment often using pairwise comparative judgements. Existing approaches typically rely on single judges or aggregate multiple judges assuming equal reliability. In practice, LLM judges vary substantially in performance across tasks and evaluation aspects, and their judgment probabilities may be biased and inconsistent. Furthermore, human-labelled supervision for judge calibration may be unavailable. We first empirically demonstrate that inconsistencies in LLM comparison probabilities exist and show that it limits the effectiveness of direct probability-based ranking. To address this, we study the LLM-asa-jury setting and propose BT-sigma, a judge-aware extension of the Bradley-Terry model that introduces a discriminator parameter for each judge to jointly infer item rankings and judge reliability from pairwise comparisons alone. Experiments on benchmark NLG evaluation datasets show that BT-sigma consistently outperforms averaging-based aggregation methods, and that the learned discriminators strongly correlate with independent measures of the cycle consistency of LLM judgments. Further analysis reveals that BT-sigma can be interpreted as an unsupervised calibration mechanism that improves aggregation by modelling judge reliability.

Mengjie Qian, Guangzhi Sun, Mark J. F. Gales +1

Department of Engineering, University of Cambridge, UK.

LLM-as-a-Judge Language Model Generation Evaluation Bradley–Terry Model Language Model Calibration

Feb 17, 2026

*-PLUIE: Personalisable metric with Llm Used for Improved Evaluation

Evaluating the quality of automatically generated text often relies on LLM-as-a-judge (LLM-judge) methods. While effective, these approaches are computationally expensive and require post-processing. To address these limitations, we build upon ParaPLUIE, a perplexity-based LLM-judge metric that estimates confidence over ``Yes/No'' answers without generating text. We introduce *-PLUIE, task specific prompting variants of ParaPLUIE and evaluate their alignment with human judgement. Our experiments show that personalised *-PLUIE achieves stronger correlations with human ratings while maintaining low computational cost.

Quentin Lemesle, Léane Jourdan, Daisy Munson +4

Univ Rennes, CNRS, IRISA, EXPRESSION, 22300 Lannion, France · Nantes Université, École Centrale Nantes, CNRS, LS2N, UMR 6004, F-44000 Nantes, France · Univ Rennes, CNRS, IRISA, SOTERN, 22300 Lannion, France +1

Language Model Generation Evaluation LLM Evaluation LLM-as-a-Judge

Feb 17, 2026

LLM-as-Judge on a Budget

LLM-as-a-judge has emerged as a cornerstone technique for evaluating large language models by leveraging LLM reasoning to score prompt-response pairs. Since LLM judgments are stochastic, practitioners commonly query each pair multiple times to estimate mean scores accurately. This raises a critical challenge: given a fixed computational budget

B

, how to optimally allocate queries across

K

prompt-response pairs to minimize estimation error? We present a principled variance-adaptive approach leveraging multi-armed bandit theory and concentration inequalities. Our method dynamically allocates queries based on estimated score variances, concentrating resources where uncertainty is highest. Further, our algorithm is shown to achieve a worst-case score-estimation error of

\tilde{O}\left(\sqrt{\frac{\sum_{i=1}^K σ_i^2}{B}}\right)

,

σ_i^2

being the unknown score variance for pair

i \in [K]

with near-optimal budget allocation. Experiments on Summarize-From-Feedback and HelpSteer2 demonstrate that our method significantly outperforms uniform allocation, reducing worst-case estimation error while maintaining identical budgets. Our work establishes a theoretical foundation for efficient LLM evaluation with practical implications for AI safety, model alignment, and automated assessment at scale.

Aadirupa Saha, Aniket Wagde, Branislav Kveton

UIC · Adobe

LLM Evaluation LLM-as-a-Judge Adaptive Sampling

Feb 16, 2026

Synthetic Reader Panels: Tournament-Based Ideation with LLM Personas for Autonomous Publishing

We present a system for autonomous book ideation that replaces human focus groups with synthetic reader panels -- diverse collections of LLM-instantiated reader personas that evaluate book concepts through structured tournament competitions. Each persona is defined by demographic attributes (age group, gender, income, education, reading level), behavioral patterns (books per year, genre preferences, discovery methods, price sensitivity), and consistency parameters. Panels are composed per imprint to reflect target demographics, with diversity constraints ensuring representation across age, reading level, and genre affinity. Book concepts compete in single-elimination, double-elimination, round-robin, or Swiss-system tournaments, judged against weighted criteria including market appeal, originality, and execution potential. To reject low-quality LLM evaluations, we implement five automated anti-slop checks (repetitive phrasing, generic framing, circular reasoning, score clustering, audience mismatch). We report results from deployment within a multi-imprint publishing operation managing 6 active imprints and 609 titles in distribution. Three case studies -- a 270-evaluator panel for a children's literacy novel, and two 5-person expert panels for a military memoir and a naval strategy monograph -- demonstrate that synthetic panels produce actionable demographic segmentation, identify structural content issues invisible to homogeneous reviewers, and enable tournament filtering that eliminates low-quality concepts while enriching high-quality survivors from 15% to 62% of the evaluated pool.

Fred Zimmerman

Nimble Books LLC / Big Five Killer LLC.

LLM-as-a-Judge User Simulation Large Language Model-Based Social Simulation

Feb 14, 2026

Rubrics as an Attack Surface: Stealthy Preference Drift in LLM Judges

Evaluation and alignment pipelines for large language models increasingly rely on LLM-based judges, whose behavior is guided by natural-language rubrics and validated on benchmarks. We identify a previously under-recognized vulnerability in this workflow, which we term Rubric-Induced Preference Drift (RIPD). Even when rubric edits pass benchmark validation, they can still produce systematic and directional shifts in a judge's preferences on target domains. Because rubrics serve as a high-level decision interface, such drift can emerge from seemingly natural, criterion-preserving edits and remain difficult to detect through aggregate benchmark metrics or limited spot-checking. We further show this vulnerability can be exploited through rubric-based preference attacks, in which benchmark-compliant rubric edits steer judgments away from a fixed human or trusted reference on target domains, systematically inducing RIPD and reducing target-domain accuracy up to 9.5% (helpfulness) and 27.9% (harmlessness). When these judgments are used to generate preference labels for downstream post-training, the induced bias propagates through alignment pipelines and becomes internalized in trained policies. This leads to persistent and systematic drift in model behavior. Overall, our findings highlight evaluation rubrics as a sensitive and manipulable control interface, revealing a system-level alignment risk that extends beyond evaluator reliability alone. The code is available at: https://github.com/ZDCSlab/Rubrics-as-an-Attack-Surface. Warning: Certain sections may contain potentially harmful content that may not be appropriate for all readers.

Ruomeng Ding, Yifei Pang, He Sun +3

University of North Carolina at Chapel Hill · 2Carnegie Mellon University · 3Yale University +1

LLM Alignment LLM-as-a-Judge Adversarial Attacks on LLMs Rubric-Based Evaluation Bias in LLM-as-a-Judge

Feb 13, 2026

CLASE: A Hybrid Method for Chinese Legalese Stylistic Evaluation

Legal text generated by large language models (LLMs) can usually achieve reasonable factual accuracy, but it frequently fails to adhere to the specialised stylistic norms and linguistic conventions of legal writing. In order to improve stylistic quality, a crucial first step is to establish a reliable evaluation method. However, having legal experts manually develop such a metric is impractical, as the implicit stylistic requirements in legal writing practice are difficult to formalise into explicit rubrics. Meanwhile, existing automatic evaluation methods also fall short: reference-based metrics conflate semantic accuracy with stylistic fidelity, and LLM-as-a-judge evaluations suffer from opacity and inconsistency. To address these challenges, we introduce CLASE (Chinese LegAlese Stylistic Evaluation), a hybrid evaluation method that focuses on the stylistic performance of legal text. The method incorporates a hybrid scoring mechanism that combines 1) linguistic feature-based scores and 2) experience-guided LLM-as-a-judge scores. Both the feature coefficients and the LLM scoring experiences are learned from contrastive pairs of authentic legal documents and their LLM-restored counterparts. This hybrid design captures both surface-level features and implicit stylistic norms in a transparent, reference-free manner. Experiments on 200 Chinese legal documents show that CLASE achieves substantially higher alignment with human judgments than traditional metrics and pure LLM-as-a-judge methods. Beyond improved alignment, CLASE provides interpretable score breakdowns and suggestions for improvements, offering a scalable and practical solution for professional stylistic evaluation in legal text generation (Code and data for CLASE is available at: https://github.com/rexera/CLASE).

Yiran Rex Ma, Yuxiao Ye, Huiyuan Xie

School of Foreign Languages, Peking University · Department of Computer Science and Technology, Tsinghua University

Automated Evaluation Legal NLP Stylometry Text Generation LLM-as-a-Judge

Feb 9, 2026

CARE: Confounder-Aware Aggregation for Reliable LLM Evaluation

LLM-as-a-judge ensembles are the standard paradigm for scalable evaluation, but their aggregation mechanisms suffer from a fundamental flaw: they implicitly assume that judges provide independent estimates of true quality. However, in practice, LLM judges exhibit correlated errors caused by shared latent confounders -- such as verbosity, stylistic preferences, or training artifacts -- causing standard aggregation rules like majority vote or averaging to provide little gain or even amplify systematic mistakes. To address this, we introduce CARE, a confounder-aware aggregation framework that explicitly models LLM judge scores as arising from both a latent true-quality signal and shared confounding factors. Rather than heuristically re-weighting judges, CARE separates quality from confounders without access to ground-truth labels. We provide theoretical guarantees for identifiability and finite-sample recovery under shared confounders, and we quantify the systematic bias incurred when aggregation models omit confounding latent factors. Across 12 public benchmarks spanning continuous scoring, binary classification, and pairwise preference settings, CARE improves aggregation accuracy, reducing error by up to 26.8%. Code is released in https://github.com/SprocketLab/CARE.

Jitian Zhao, Changho Shin, Tzu-Heng Huang +2

Department of Statistics, University of Wisconsin–Madison, Madison, WI, USA · Department of Computer Sciences, University of Wisconsin–Madison, Madison, WI, USA.

LLM Evaluation LLM-as-a-Judge Language Model Generation Evaluation Language Model Ensembles

Feb 6, 2026

FairJudge: An Adaptive, Debiased, and Consistent LLM-as-a-Judge

Existing LLM-as-a-Judge systems suffer from three fundamental limitations: limited adaptivity to task- and domain-specific evaluation criteria, systematic biases driven by non-semantic cues such as position, length, format, and model provenance, and evaluation inconsistency that leads to contradictory judgments across different evaluation modes (e.g., pointwise versus pairwise). To address these issues, we propose FairJudge, an adaptive, debiased, and consistent LLM-as-a-Judge. Unlike prior approaches that treat the judge as a static evaluator, FairJudge models judging behavior itself as a learnable and regularized policy. From a data-centric perspective, we construct a high-information-density judging dataset that explicitly injects supervision signals aligned with evaluation behavior. Building on this dataset, we adopt a curriculum-style SFT-DPO-GRPO training paradigm that progressively aligns rubric adherence, bias mitigation, and cross-mode consistency, while avoiding catastrophic forgetting. Experimental results on multiple internal and public benchmarks show that FairJudge consistently improves agreement and F1, reduces non-semantic biases, and outperforms substantially larger instruction-tuned LLMs. All resources will be publicly released after acceptance to facilitate future research.

Bo Yang, Lanfei Feng, Yunkui Chen +3

College of Computer Science, Zhejiang University, Hangzhou, China · School of Software Technology, Zhejiang University, Hangzhou, China

LLM Evaluation LLM-as-a-Judge Language Model Bias Evaluation Bias in LLM-as-a-Judge

Feb 6, 2026

JADE: Expert-Grounded Dynamic Evaluation for Open-Ended Professional Tasks

Evaluating agentic AI on open-ended professional tasks faces a fundamental dilemma between rigor and flexibility. Static rubrics provide rigorous, reproducible assessment but fail to accommodate diverse valid response strategies, while LLM-as-a-judge approaches adapt to individual responses yet suffer from instability and bias. Human experts address this dilemma by combining domain-grounded principles with dynamic, claim-level assessment. Inspired by this process, we propose JADE, a two-layer evaluation framework. Layer 1 encodes expert knowledge as a predefined set of evaluation skills, providing stable evaluation criteria. Layer 2 performs report-specific, claim-level evaluation to flexibly assess diverse reasoning strategies, with evidence-dependency gating to invalidate conclusions built on refuted claims. Experiments on BizBench show that JADE improves evaluation stability and reveals critical agent failure modes missed by holistic LLM-based evaluators. We further demonstrate strong alignment with expert-authored rubrics and effective transfer to HealthBench and DR.BENCH, covering medical and 10-domain professional evaluation settings. Code and data are available at https://github.com/smiling-world/JADE.

Lanbo Lin, Jiayao Liu, Tianyuan Yang +5

Alibaba International Digital Commerce Group · Zhejiang University · Peking University

LLM-as-a-Judge AI Agent Evaluation Rubric-Based Evaluation Automated Agent Evaluation

Feb 3, 2026

MAS-ProVe: Understanding the Process Verification of Multi-Agent Systems

Multi-Agent Systems (MAS) built on Large Language Models (LLMs) often exhibit high variance in their reasoning trajectories. Process verification, which evaluates intermediate steps in trajectories, has shown promise in general reasoning settings, and has been suggested as a potential tool for guiding coordination of MAS; however, its actual effectiveness in MAS remains unclear. To fill this gap, we present MAS-ProVe, a systematic empirical study of process verification for multi-agent systems (MAS). Our study spans three verification paradigms (LLM-as-a-Judge, reward models, and process reward models), evaluated across two levels of verification granularity (agent-level and iteration-level). We further examine five representative verifiers and four context management strategies, and conduct experiments over six diverse MAS frameworks on multiple reasoning benchmarks. We find that process-level verification does not consistently improve performance and frequently exhibits high variance, highlighting the difficulty of reliably evaluating partial multi-agent trajectories. Among the methods studied, LLM-as-a-Judge generally outperforms reward-based approaches, with trained judges surpassing general-purpose LLMs. We further observe a small performance gap between LLMs acting as judges and as single agents, and identify a context-length-performance trade-off in verification. Overall, our results suggest that effective and robust process verification for MAS remains an open challenge, requiring further advances beyond current paradigms. Code is available at https://github.com/Wang-ML-Lab/MAS-ProVe.

Vishal Venkataramani, Haizhou Shi, Zixuan Ke +6

1Rutgers University · 2Salesforce AI Research

Multi-Agent LLM Systems LLM-as-a-Judge Process Reward Models LLM Agent Evaluation Multi-Agent Systems