mlsanityData to 7 Oct 2026

Evaluating the Consistency of LLM Evaluators

arXiv 2412.00543 ↗First posted 30 Nov 2024cs.CLcited by 24

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) have shown potential as general evaluators along with the evident benefits of speed and cost. While their correlation against human annotators has been widely studied, consistency as evaluators is still understudied, raising concerns about the reliability of LLM evaluators. In this paper, we conduct extensive studies on the two aspects of consistency in LLM evaluations, Self-Consistency (SC) and Inter-scale Consistency (IC), on different scoring scales and criterion granularity with open-source and proprietary models. Our comprehensive analysis demonstrates that strong proprietary models are not necessarily consistent evaluators, highlighting the importance of considering consistency in assessing the capability of LLM evaluators.

Key facts. Cited by 24 papers in the map (data through 7 Oct 2026). Builds on 14 of its 26 references that are in the map. First posted on arXiv 30 Nov 2024.

Builds on

14 of 26 references.

References · 14

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  3. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,634
  4. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,919
  5. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,205
  6. 2023LIMA: Less Is More for Alignmentcited by 1,080
  7. 2023ChatGPT Outperforms Crowd-Workers for Text-Annotation Taskscited by 849
  8. 2023Can Large Language Models Be an Alternative to Human Evaluations?cited by 729
  9. 2023Prometheus: Inducing Fine-grained Evaluation Capability in Language Modelscited by 558
  10. 2023Zephyr: Direct Distillation of LM Alignmentcited by 498
  11. 2024Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Modelscited by 431
  12. 2023Camels in a Changing Climate: Enhancing LM Adaptation with Tulu 2cited by 227
2 more
  1. 2023FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Setscited by 198
  2. 2024Large Language Models are Inconsistent and Biased Evaluatorscited by 146

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 17; 2026: 7.

24 papers.

Sort:

By area:

  1. 2025Improving LLM-as-a-Judge Inference with the Judgment Distributioncited by 36
  2. 2025A Technical Survey of Reinforcement Learning Techniques for Large Language Modelscited by 19
  3. 2026Grading Scale Impact on LLM-as-a-Judge: Human-LLM Alignment Is Highest on 0-5 Grading Scalecited by 17
  4. 2025Media and responsible AI governance: a game-theoretic and LLM analysiscited by 11
  5. 2025Do LLMs trust AI regulation? Emerging behaviour of game-theoretic LLM agentscited by 11
  6. 2025Taking Notes Brings Focus? Towards Multi-Turn Multimodal Dialogue Learningcited by 5
  7. 2025CAPE: Context-Aware Personality Evaluation Framework for Large Language Modelscited by 4
  8. 2025The Non-Determinism of Small LLMs: Evidence of Low Answer Consistency in Repetition Trials of Standard Multiple-Choice Benchmarkscited by 3
  9. 2025DialogGuard: Multi-Agent Psychosocial Safety Evaluation of Sensitive LLM Responsescited by 3
  10. 2025HopWeaver: Cross-Document Synthesis of High-Quality and Authentic Multi-Hop Questionscited by 2
  11. 2025AraTable: Benchmarking LLMs' Reasoning and Understanding of Arabic Tabular Datacited by 2
  12. 2025A Dual-Perspective NLG Meta-Evaluation Framework with Automatic Benchmark and Better Interpretabilitycited by 1
12 more
  1. 2025CancerGUIDE: Cancer Guideline Understanding via Internal Disagreement Estimationcited by 1
  2. 2025CAT: A Metric-Driven Framework for Analyzing the Consistency-Accuracy Relation of LLMs under Controlled Input Variationscited by 1
  3. 2026Judge Circuits Explain Format-Induced Inconsistency in LLM-as-a-Judgecited by 1
  4. 2026Generative-Evaluative Agreement: A Necessary Validity Criterion for LLM-Enabled Adaptive Assessmentcited by 1
  5. 2025Surfacing Semantic Orthogonality Across Model Safety Benchmarks: A Multi-Dimensional Analysiscited by 0
  6. 2025How Good LLM-Generated Password Policies Are?cited by 0
  7. 2025A Comprehensive Evaluation framework of Alignment Techniques for LLMscited by 0
  8. 2025Harmonizing Diverse Models: A Layer-wise Merging Strategy for Consistent Generationcited by 0
  9. 2026What Are We Measuring in NLG? A Meta-Analysis of Evaluation Trends 2020-2025cited by 0
  10. 2026MCJudgeBench: A Benchmark for Constraint-Level Judge Evaluation in Multi-Constraint Instruction Followingcited by 0
  11. 2026Agent Step Value: Auditing Evaluator-Channel Reversals in Black-Box Agent Tracescited by 0
  12. 2026Query Timing Produces Opposite Positional Biases Between LLMs and Humanscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Exploring the Reliability of Large Language Models as Customized Evaluators for Diverse NLP Tasks
  2. 2024Can Large Language Models be Trusted for Evaluation? Scalable Meta-Evaluation of LLMs as Evaluators via Agent Debate
  3. 2025Neither Valid nor Reliable? Investigating the Use of LLMs as Judges
  4. 2025On Evaluating LLM Alignment by Evaluating LLMs as Judges
  5. 2026When Consistency Does Not Mean Reliability: Evaluating Local LLM Judges Against Human Ratings
  6. 2025ConsistencyChecker: Tree-based Evaluation of LLM Generalization Capabilities
  7. 2024METAL: Towards Multilingual Meta-Evaluation
  8. 2024Enhancing LLM Evaluations: The Garbling Trick