Lune

ACL2023Top-tier venue

Can Large Language Models Be an Alternative to Human Evaluations?

David Cheng-Han Chiang, Hung-yi Lee

Abstract

Human evaluation is indispensable and inevitable for assessing the quality of texts generated by machine learning models or written by humans. However, human evaluation is very difficult to reproduce and its quality is notoriously unstable, hindering fair comparisons among different natural language processing (NLP) models and algorithms. Recently, large language models (LLMs) have demonstrated exceptional performance on unseen tasks when only the task instructions are provided. In this paper, we explore if such an ability of the LLMs can be used as an alternative to human evaluation. We present the LLMs with the exact same instructions, samples to be evaluated, and questions used to conduct human evaluation, and then ask the LLMs to generate responses to those questions; we dub this LLM evaluation. We use human evaluation and LLM evaluation to evaluate the texts in two NLP tasks: open-ended story generation and adversarial attacks. We show that the result of LLM evaluation is consistent with the results obtained by expert human evaluation: the texts rated higher by human experts are also rated higher by the LLMs. We also find that the results of LLM evaluation are stable over different formatting of the task instructions and the sampling algorithm used to generate the answer. We are the first to show the potential of using LLMs to assess the quality of texts and discuss the limitations and ethical considerations of LLM evaluation.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

151 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher cites Training language models to follow instructions with human feedback
  • GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher cites Safe RLHF: Safe Reinforcement Learning from Human Feedback
  • Multitask Prompted Training Enables Zero-Shot Task Generalization cites Language Models are Few-Shot Learners
  • Multitask Prompted Training Enables Zero-Shot Task Generalization cites Finetuned Language Models are Zero-Shot Learners
  • Finetuned Language Models are Zero-Shot Learners cites Multitask Prompted Training Enables Zero-Shot Task Generalization
  • Language Models are Few-Shot Learners cites The Curious Case of Neural Text Degeneration
  • Finetuned Language Models are Zero-Shot Learners cites Language Models are Few-Shot Learners
  • Finetuned Language Models are Zero-Shot Learners cites Training language models to follow instructions with human feedback
  • Training language models to follow instructions with human feedback cites Finetuned Language Models are Zero-Shot Learners
  • Training language models to follow instructions with human feedback cites Multitask Prompted Training Enables Zero-Shot Task Generalization
  • Training language models to follow instructions with human feedback cites Language Models are Few-Shot Learners
  • ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate cites Multitask Prompted Training Enables Zero-Shot Task Generalization
  • ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate cites Language Models are Few-Shot Learners
  • ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate cites Finetuned Language Models are Zero-Shot Learners
  • ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate cites Training language models to follow instructions with human feedback
  • Safe RLHF: Safe Reinforcement Learning from Human Feedback cites Multitask Prompted Training Enables Zero-Shot Task Generalization
  • Safe RLHF: Safe Reinforcement Learning from Human Feedback cites Language Models are Few-Shot Learners
  • Safe RLHF: Safe Reinforcement Learning from Human Feedback cites Training language models to follow instructions with human feedback
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers151

  1. Chatbot Arena: An Open Platform for Evaluating LLMs by Human PreferenceWei-Lin Chiang, Lianmin Zheng, Ying Sheng, Anastasios Nikolas Angelopoulos et al.ICML 2024 · 1,212 citations
  2. ChatEval: Towards Better LLM-based Evaluators through Multi-Agent DebateChi-Min Chan, Weize Chen, Yusheng Su, Jianxuan Yu et al.ICLR 2024 · 871 citations
  3. Safe RLHF: Safe Reinforcement Learning from Human FeedbackJosef Dai, Xuehai Pan, Ruiyang Sun, Jiaming Ji et al.ICLR 2024 · 656 citations
  4. Prometheus: Inducing Fine-Grained Evaluation Capability in Language ModelsSeungone Kim, Jamin Shin, Yejin Choi, Joel Jang et al.ICLR 2024 · 468 citations
  5. GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via CipherYouliang Yuan, Wenxiang Jiao, Wenxuan Wang, Jen-tse Huang et al.ICLR 2024 · 441 citations

Builds on15

  1. Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
  2. Training language models to follow instructions with human feedbackLong Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida et al.NeurIPS 2022 · 24,707 citations
  3. Finetuned Language Models are Zero-Shot LearnersJason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu et al.ICLR 2022 · 4,966 citations
  4. The Curious Case of Neural Text DegenerationAri Holtzman, Jan Buys, Li Du, Maxwell Forbes et al.ICLR 2020 · 4,112 citations
  5. Multitask Prompted Training Enables Zero-Shot Task GeneralizationVictor Sanh, Albert Webson, Colin Raffel, Stephen H. Bach et al.ICLR 2022 · 1,976 citations

Related papers

  1. Rating Roulette: Self-Inconsistency in LLM-As-A-Judge FrameworksRajarshi Haldar, Julia HockenmaierEMNLP Findings 2025 · 11 citations
  2. Themis: A Reference-free NLG Evaluation Language Model with Flexibility and InterpretabilityXinyu Hu, Li Lin, Mingqi Gao, Xunjian Yin et al.EMNLP 2024 · 2 citations
  3. A Closer Look into Using Large Language Models for Automatic EvaluationDavid Cheng-Han Chiang, Hung-yi LeeEMNLP Findings 2023 · 33 citations
  4. Unveiling the Achilles' Heel of NLG Evaluators: A Unified Adversarial Framework Driven by Large Language ModelsYiming Chen, Chen Zhang, Danqing Luo, Luis Fernando D'Haro et al.ACL Findings 2024
  5. Systematic Task Exploration with LLMs: A Study in Citation Text GenerationFurkan Sahinuç, Ilia Kuznetsov, Yufang Hou, Iryna GurevychACL 2024
Dusk over the sea between two cliffs drawn in fine vertical lines