mlsanityData to 14 Oct 2026

Are LLM-based Evaluators Confusing NLG Quality Criteria?

arXiv 2402.12055 ↗First posted 19 Feb 2024cs.CLcited by 36

Not on the map. Why?

Abstract (arXiv)

Some prior work has shown that LLMs perform well in NLG evaluation for different tasks. However, we discover that LLMs seem to confuse different evaluation criteria, which reduces their reliability. For further verification, we first consider avoiding issues of inconsistent conceptualization and vague expression in existing NLG quality criteria themselves. So we summarize a clear hierarchical classification system for 11 common aspects with corresponding different criteria from previous studies involved. Inspired by behavioral testing, we elaborately design 18 types of aspect-targeted perturbation attacks for fine-grained analysis of the evaluation behaviors of different LLMs. We also conduct human annotations beyond the guidance of the classification system to validate the impact of the perturbations. Our experimental results reveal confusion issues inherent in LLMs, as well as other noteworthy phenomena, and necessitate further research and improvements for LLM-based evaluation.

Key facts. Cited by 36 papers in the map (data through 14 Oct 2026). Builds on 31 of its 49 references that are in the map. First posted on arXiv 19 Feb 2024.

Builds on

31 of 49 references.

References · 31

  1. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  2. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  3. 2019BERTScore: Evaluating Text Generation with BERTcited by 5,512
  4. 2023G-Eval: NLG Evaluation using GPT-4 with Better Human Alignmentcited by 2,276
  5. 2020BLEURT: Learning Robust Metrics for Text Generationcited by 1,203
  6. 2023Large Language Models are not Fair Evaluatorscited by 1,085
  7. 2015Teaching Machines to Read and Comprehendcited by 1,049
  8. 2020COMET: A Neural Framework for MT Evaluationcited by 1,033
  9. 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 1,006
  10. 2023Can Large Language Models Be an Alternative to Human Evaluations?cited by 734
  11. 2020SummEval: Re-evaluating Summarization Evaluationcited by 674
  12. 2019Evaluating the Factual Consistency of Abstractive Text Summarizationcited by 584
19 more
  1. 2023Prometheus: Inducing Fine-grained Evaluation Capability in Language Modelscited by 575
  2. 2019SAMSum Corpus: A Human-annotated Dialogue Dataset for Abstractive Summarizationcited by 535
  3. 2023Is ChatGPT a Good NLG Evaluator? A Preliminary Studycited by 422
  4. 2023Large Language Models Are State-of-the-Art Evaluators of Translation Qualitycited by 420
  5. 2019Neural Text Summarization: A Critical Evaluationcited by 189
  6. 2020USR: An Unsupervised and Reference Free Evaluation Metric for Dialog Generationcited by 188
  7. 2023Human-like Summarization Evaluation with ChatGPTcited by 126
  8. 2023Are Large Language Model-based Evaluators the Solution to Scaling Up Multilingual Evaluation?cited by 88
  9. 2023EvalLM: Interactive Evaluation of Large Language Model Prompts on User-Defined Criteriacited by 74
  10. 2023The Next Chapter: A Study of Large Language Models in Storytellingcited by 58
  11. 2023Evaluation Metrics in the Era of GPT-4: Reliably Evaluating Large Language Models on Sequence to Sequence Taskscited by 53
  12. 2021Perturbation CheckLists for Evaluating NLG Evaluation Metricscited by 50
  13. 2022On the Blind Spots of Model-Based Evaluation Metrics for Text Generationcited by 49
  14. 2023A Comprehensive Analysis of the Effectiveness of Large Language Models as Automatic Dialogue Evaluatorscited by 45
  15. 2022DEMETR: Diagnosing Evaluation Metrics for Translationcited by 36
  16. 2023X-Eval: Generalizable Multi-aspect Text Evaluation via Augmented Instruction Tuning with Auxiliary Evaluation Aspectscited by 28
  17. 2023Simple LLM Prompting is State-of-the-Art for Robust and Multilingual Dialogue Evaluationcited by 19
  18. 2023CoAScore: Chain-of-Aspects Prompting for NLG Evaluationcited by 14
  19. 2023DeltaScore: Fine-Grained Story Evaluation with Perturbationscited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 8; 2025: 18; 2026: 10.

36 papers.

Sort:

By area:

  1. 2024From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judgecited by 494
  2. 2024Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judgescited by 205
  3. 2024Lessons from the Trenches on Reproducible Evaluation of Language Modelscited by 179
  4. 2025Neither Valid nor Reliable? Investigating the Use of LLMs as Judgescited by 39
  5. 2025Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarkscited by 28
  6. 2024Improving Context-Aware Preference Modeling for Language Modelscited by 22
  7. 2024Reasoning Through Execution: Unifying Process and Outcome Rewards for Code Generationcited by 21
  8. 2024RevisEval: Improving LLM-as-a-Judge via Response-Adapted Referencescited by 20
  9. 2024Themis: A Reference-free NLG Evaluation Language Model with Flexibility and Interpretabilitycited by 19
  10. 2025Beyond the Surface: Enhancing LLM-as-a-Judge Alignment with Human via Internal Representationscited by 15
  11. 2024DHP Benchmark: Are LLMs Good NLG Evaluators?cited by 12
  12. 2025Decision Information Meets Large Language Models: The Future of Explainable Operations Researchcited by 9
24 more
  1. 2025Rubrik's Cube: Testing a New Rubric for Evaluating Explanations on the CUBE datasetcited by 8
  2. 2025Multi-Agent LLM Judge: automatic personalized LLM judge design for evaluating natural language generation applicationscited by 7
  3. 2025CLEV: LLM-Based Evaluation Through Lightweight Efficient Voting for Free-Form Question-Answeringcited by 6
  4. 2025HPSS: Heuristic Prompting Strategy Search for LLM Evaluatorscited by 5
  5. 2025Exploring the Multilingual NLG Evaluation Abilities of LLM-Based Evaluatorscited by 5
  6. 2025Auto-Prompt Ensemble for LLM Judgecited by 5
  7. 2025OpeNLGauge: An Explainable Metric for NLG Evaluation with Open-Weights LLMscited by 4
  8. 2025Revisiting NLI: Towards Cost-Effective and Human-Aligned Metrics for Evaluating LLMs in Question Answeringcited by 4
  9. 2026Focal Reward: Balanced Reinforcement Learning under Rubric-Based Rewardscited by 4
  10. 2026Meta-Judging with Large Language Models: Concepts, Methods, and Challengescited by 3
  11. 2026Useless but Safe? Benchmarking Utility Recovery with User Intent Clarification in Multi-Turn Conversationscited by 3
  12. 2025EvolvR: Self-Evolving Pairwise Reasoning for Story Evaluation to Enhance Generationcited by 2
  13. 2025Calibrating MLLM-as-a-judge via Multimodal Bayesian Prompt Ensemblescited by 2
  14. 2025Reward Models are Metrics in a Trench Coatcited by 2
  15. 2026Reflections and New Directions for Human-Centered Large Language Modelscited by 2
  16. 2025A Dual-Perspective NLG Meta-Evaluation Framework with Automatic Benchmark and Better Interpretabilitycited by 1
  17. 2026What Are We Measuring in NLG? A Meta-Analysis of Evaluation Trends 2020-2025cited by 1
  18. 2026Graph-Structured Rubrics: Compiling Rubrics into Typed Evaluation Graphs for LLM Judgescited by 1
  19. 2025Transparent Reference-free Automated Evaluation of Open-Ended User Survey Responsescited by 0
  20. 2025MTQ-Eval: Multilingual Text Quality Evaluation for Language Modelscited by 0
  21. 2026SAD: A Large-Scale Strategic Argumentative Dialogue Datasetcited by 0
  22. 2026C3-Bench: A Context-Aware Change Captioning Benchmarkcited by 0
  23. 2026Beyond Scores: Understanding LLM-as-a-Judge Mechanisms in Summarization Evaluationcited by 0
  24. 2026LLJ Cards: Best practices for the Use of LLMs as Judgescited by 0