Elamur Research⌕
☰
home / papers / 2024 / TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generation

TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generation

Paper · 2024 · Jonathan Cook, Tim Rocktaschel, Jakob Foerster et al.

Summary

Judging how well LLMs follow instructions usually relies on pairwise preferences or single holistic scores. These hide the reasoning behind a verdict and often produce disagreement, both among humans and between humans and LLM judges. As an alternative, the authors introduce an evaluation protocol that requires no human involvement, which they call TICK (Targeted Instruct-evaluation with ChecKlists). An LLM first breaks each instruction into a tailored checklist of YES/NO questions, one per requirement, whether stated outright or implicit. A judge model then answers each question about a candidate response, and the answers are combined into a Pass Rate per response and DRFR across a dataset. On an internal set of 612 annotator-written instructions, the generated checklists resemble human-written ones about as closely as a second human checklist does. TICK brings LLM judgments closer to human preferences than direct scoring. The same checklists also drive STICK (Self-TICK), which uses self-generated checklists for in-context refinement and Best-of-N selection. Because the method needs no handcrafted checklists, it offers a cheap and interpretable way to both evaluate models and improve their outputs on arbitrary instruction datasets.

Key points

  • Agreement with humans: exact agreement between LLM judgments and human preferences rose from 46.4% to 52.2% with TICK, compared with having the LLM score outputs directly.
  • Checklist quality: GPT-4o checklists reached BLEU 0.759 against ground-truth human checklists (H*). An alternate human-written set (H′) scored 0.733.
  • Self-refinement on LiveBench: STICK raised Command-R+ overall from 32.0 to 35.8, including +7.8 on reasoning. Unstructured Self-Refine instead dropped it to 23.7. GPT-4o gained 0.8 with STICK.
  • Other benchmarks: STICK improved Command-R+ by 6.5% on InFoBench and 7.1% on WildBench. For Best-of-N selection, it also beat a general-purpose reward model.
  • Human annotation: giving annotators generated checklists raised inter-annotator agreement on WildBench from 0.194 to 0.256. 78.5% of annotators said the checklist made scoring easier, though the authors caution that it should serve only as a partial guide.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓University of OxfordOxford · United Kingdom138 works · 161,448 citations🎓University College LondonLondon · United Kingdom119 works · 23,545 citations

 Fields, topics and concepts

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 19,503 worksArtificial Intelligence · 18,805 worksArts and Humanities · 17,959 worksLanguage and Linguistics · 17,789 worksComputer Vision and Pattern Recognition · 3,621 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 4,145 works

Cite

Cite as Jonathan Cook, Tim Rocktaschel, Jakob Foerster, Dennis Aumiller and Alex Wang. 2024. TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generation. arXiv:2410.03608.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2410.03608 · arXiv 2410.03608

@misc{cook2024ticking,
  title = {{TICKing All the Boxes: Generated Checklists Improve LLM Evaluation and Generation}},
  author = {Jonathan Cook, Tim Rocktaschel, Jakob Foerster, Dennis Aumiller and Alex Wang},
  year = {2024},
  doi = {10.48550/arxiv.2410.03608},
  eprint = {2410.03608},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2410.03608}
}

References

50 references, 29 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Zachary Ankner, Mansheej Paul, Brandon Cui, Jonathan D Chang, and Prithviraj Ammanabrolu. Critique-out-Loud Reward Models. arXiv preprint arXiv:2408.11791, 2024.
  2. Anthropic. The Claude 3 Model Family: Opus, Sonnet, Haiku, 2023. URL https://www-cdn.anthropic.com/de8ba9b01c9ab7cbabf5c33b80b7bbc618857627/Model_ Card_Claude_3.pdfs. ↗
  3. Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan, Nicholas Joseph, Saurav Kadavath, Jackson Kernion, Tom Conerly, Sheer El Showk, Nelson Elhage, Zac Hatfield-Dodds, Danny Hernandez, Tristan Hume, Scott Johnston, Sh… find
  4. Yuntao Bai, Saurav Kadavath, Sandipan Kundu, Amanda Askell, Jackson Kernion, Andy Jones, Anna Chen, Anna Goldie, Azalia Mirhoseini, Cameron McKinnon, Carol Chen, Catherine Olsson, Christopher Olah, Danny Hernandez, Dawn Drain, Deep Ganguli, Dustin Li, Eli Tran-Johnson, Ethan Perez, Jamie Kerr, Jared… — Constitutional AI: Harmlessness from AI Feedback
  5. Yew Ken Chia, Pengfei Hong, Lidong Bing, and Soujanya Poria. INSTRUCTEVAL: Towards Holistic Evaluation of Instruction-Tuned Large Language Models. arXiv preprint arXiv:2306.04757, 2023.
  6. Cheng-Han Chiang and Hung-yi Lee. A closer look into using large language models for automatic evaluation. In Houda Bouamor, Juan Pino, and Kalika Bali (eds.), Findings of the Association for Computational Linguistics: EMNLP 2023, pp. 8928–8942, Singapore, December 2023. Association for Computationa… find
  7. Cohere. Command R+ Model, 2024. command-r-plus. find
  8. URL — The Llama 3 Herd of Models https://docs.cohere.com/docs/Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. The Llama 3 Herd of Models. arXiv preprint arXiv:2407.21783, 2024. ↗
  9. Amelia Glaese, Nat McAleese, Maja Trebacz, John Aslanides, Vlad Firoiu, Timo Ewalds, Maribeth Rauh, Laura Weidinger, Martin J. Chadwick, Phoebe Thacker, Lucy Campbell-Gillingham, Jonathan Uesato, Po-Sen Huang, Ramona Comanescu, Fan Yang, Abigail See, Sumanth Dathathri, Rory Greig, Charlie Chen, Doug… — Improving alignment of dialogue agents via targeted human judgements
  10. Tom Hosking, Phil Blunsom, and Max Bartolo. Human Feedback is not Gold Standard. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024. find
  11. OpenReview.net, 2024. URL https://openreview.net/forum?id= 7W3GLNImfS. ↗
  12. Jie Huang, Xinyun Chen, Swaroop Mishra, Huaixiu Steven Zheng, Adams Wei Yu, Xinying Song, and Denny Zhou. Large language models cannot self-correct reasoning yet. In The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024.
  13. OpenReview.net, 2024. URL https://openreview.net/forum?id=IkmD3fKBPQ. ↗
  14. Yuxin Jiang, Yufei Wang, Xingshan Zeng, Wanjun Zhong, Liangyou Li, Fei Mi, Lifeng Shang, Xin Jiang, Qun Liu, and Wei Wang. FollowBench: A Multi-level Fine-grained Constraints Following Benchmark for Large Language Models. In Lun-Wei Ku, Andre Martins, and Vivek Srikumar (eds.), Proceedings of the 62…
  15. Ryo Kamoi, Yusen Zhang, Nan Zhang, Jiawei Han, and Rui Zhang. When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs. arXiv preprint arXiv:2406.01297, 2024.
  16. Klaus Krippendorff. Content Analysis: An Introduction to Its Methodology. SAGE Publications, 1980. find
  17. Patrick S. H. Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks. Advances in Neural Information Proce…
  18. Xuechen Li, Tianyi Zhang, Yann Dubois, Rohan Taori, Ishaan Gulrajani, Carlos Guestrin, Percy Liang, and Tatsunori B. Hashimoto. AlpacaEval: An Automatic Evaluator of Instruction-following Models, 2023. URL https://github.com/tatsu-lab/alpaca_eval. ↗
  19. Bill Yuchen Lin, Yuntian Deng, Khyathi Chandu, Faeze Brahman, Abhilasha Ravichander, Valentina Pyatkin, Nouha Dziri, Ronan Le Bras, and Yejin Choi. WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wild. arXiv preprint arXiv:2406.04770, 2024.
  20. Chin-Yew Lin. ROUGE: A Package for Automatic Evaluation of Summaries. In Text Summarization Branches Out, pp. 74–81, Barcelona, Spain, July 2004. Association for Computational Linguistics. URL https://aclanthology.org/W04-1013. ↗
  21. Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, Shashank Gupta, Bodhisattwa Prasad Majumder, Katherine Hermann, Sean Welleck, Amir Yazdanbakhsh, and Peter Clark. Self-Refine: Iterative Refinement with Self-… find
  22. Nat McAleese, Rai Michael Pokorny, Juan Felipe Ceron Uribe, Evgenia Nitishinskaya, Maja Trebacz, and Jan Leike. LLM Critics Help Catch LLM Bugs. arXiv preprint arXiv:2407.00215, 2024. — [arXiv:2407.00215](https://arxiv.org/abs/2407.00215) find
  23. Swaroop Mishra, Daniel Khashabi, Chitta Baral, and Hannaneh Hajishirzi. Cross-Task Generalization via Natural Language Crowdsourcing Instructions. In Smaranda Muresan, Preslav Nakov, and Aline Villavicencio (eds.), Proceedings of the 60th Annual Meeting of the Association for Computational Linguisti… find
  24. Jekaterina Novikova, Ondřej Dušek, and Verena Rieser. RankME: Reliable Human Ratings for Natural Language Generation. In Marilyn Walker, Heng Ji, and Amanda Stent (eds.), Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua…
  25. OpenAI. GPT-4 Technical Report. arXiv preprint arXiv:2303.08774, 2023.
  26. OpenAI. Hello GPT-4o, 2024. URL https://openai.com/index/hello-gpt-4o/. ↗
  27. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul F. Christiano, Jan Leike, and Ryan Lowe. Train… — Training language models to follow instructions with human feedback
  28. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. Bleu: a Method for Automatic Evaluation of Machine Translation. In Pierre Isabelle, Eugene Charniak, and Dekang Lin (eds.), Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pp. 311–318, Philadelphia,…
  29. Yiwei Qin, Kaiqiang Song, Yebowen Hu, Wenlin Yao, Sangwoo Cho, Xiaoyang Wang, Xuansheng Wu, Fei Liu, Pengfei Liu, and Dong Yu. InFoBench: Evaluating Instruction Following Ability in Large Language Models. In Lun-Wei Ku, Andre Martins, and Vivek Srikumar (eds.), Findings of the Association for Comput…
  30. William Saunders, Catherine Yeh, Jeff Wu, Steven Bills, Long Ouyang, Jonathan Ward, and Jan Leike. Self-critiquing models for assisting human evaluators. arXiv preprint arXiv:2206.05802, 2022.
  31. Jérémy Scheurer, Jon Ander Campos, Tomasz Korbak, Jun Shern Chan, Angelica Chen, Kyunghyun Cho, and Ethan Perez. Training Language Models with Language Feedback at Scale. arXiv preprint arXiv:2303.16755, 2023. find
  32. Charlie Snell, Jaehoon Lee, Kelvin Xu, and Aviral Kumar. Scaling LLM test-time compute optimally can be more effective than scaling model parameters. CoRR, abs/2408.03314, 2024. doi: 10. 48550/ARXIV.2408.03314. URL https://doi.org/10.48550/arXiv.2408.03314. ↗
  33. Haoran Sun, Lixin Liu, Junjie Li, Fengyu Wang, Baohua Dong, Ran Lin, and Ruohui Huang. Conifer: Improving Complex Constrained Instruction-Following Ability of Large Language Models. arXiv preprint arXiv:2404.02823, 2024.
  34. Ye Tian, Baolin Peng, Linfeng Song, Lifeng Jin, Dian Yu, Haitao Mi, and Dong Yu. Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing. arXiv preprint arXiv:2404.12253, 2024.
  35. Gladys Tyen, Hassan Mansoor, Victor Carbune, Peter Chen, and Tony Mak. LLMs cannot find reasoning errors, but can correct them given the error location. In Lun-Wei Ku, Andre Martins, and Vivek Srikumar (eds.), Findings of the Association for Computational Linguistics ACL 2024, pp. 13894–13908, Bangk…
  36. Pat Verga, Sebastian Hofstatter, Sophia Althammer, Yixuan Su, Aleksandra Piktus, Arkady Arkhangorodsky, Minjie Xu, Naomi White, and Patrick Lewis. Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models. arXiv preprint arXiv:2404.18796, 2024.
  37. Tu Vu, Kalpesh Krishna, Salaheddin Alzubi, Chris Tar, Manaal Faruqui, and Yun-Hsuan Sung. Foundational Autoraters: Taming Large Language Models for Better Automatic Evaluation. arXiv preprint arXiv:2407.10817, 2024. find
  38. Haoxiang Wang, Wei Xiong, Tengyang Xie, Han Zhao, and Tong Zhang. Interpretable Preferences via Multi-Objective Reward Modeling and Mixture-of-Experts. arXiv preprint arXiv:2406.12845, 2024a.
  39. Tianlu Wang, Ping Yu, Xiaoqing Ellen Tan, Sean O’Brien, Ramakanth Pasunuru, Jane Dwivedi-Yu, Olga Golovneva, Luke Zettlemoyer, Maryam Fazel-Zarandi, and Asli Celikyilmaz. Shepherd: A critic for language model generation. arXiv preprint arXiv:2308.04592, 2023a.
  40. Tianlu Wang, Ilia Kulikov, Olga Golovneva, Ping Yu, Weizhe Yuan, Jane Dwivedi-Yu, Richard Yuanzhe Pang, Maryam Fazel-Zarandi, Jason Weston, and Xian Li. Self-Taught Evaluators. arXiv preprint arXiv:2408.02666, 2024b.
  41. Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A. Smith, Daniel Khashabi, and Hannaneh Hajishirzi. Self-Instruct: Aligning Language Models with Self-Generated Instructions. In Anna Rogers, Jordan Boyd-Graber, and Naoaki Okazaki (eds.), Proceedings of the 61st Annual Meeting of the Asso… find
  42. Jason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M. Dai, and Quoc V. Le. Finetuned Language Models are Zero-Shot Learners. In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022.
  43. OpenReview.net, 2022a. URL https://openreview.net/forum? id=gEZrGCozdqR. ↗
  44. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H. Chi, Quoc V. Le, and Denny Zhou. Chain-of-Thought Prompting Elicits Reasoning in Large Language Models. In Sanmi Koyejo, S. Mohamed, A. Agarwal, Danielle Belgrave, K. Cho, and A. Oh (eds.), Advances in Neural Inform…
  45. Bosi Wen, Pei Ke, Xiaotao Gu, Lindong Wu, Hao Huang, Jinfeng Zhou, Wenchuang Li, Binxin Hu, Wendy Gao, Jiaxin Xu, Yiming Liu, Jie Tang, Hongning Wang, and Minlie Huang. Benchmarking Complex Instruction-Following with Multiple Constraints Composition. arXiv preprint arXiv:2407.03978, 2024. — [arXiv:2407.03978](https://arxiv.org/abs/2407.03978) find
  46. Colin White, Samuel Dooley, Manley Roberts, Arka Pal, Benjamin Feuer, Siddhartha Jain, Ravid Shwartz-Ziv, Neel Jain, Khalid Saifullah, Siddartha Naidu, Chinmay Hegde, Yann LeCun, Tom Goldstein, Willie Neiswanger, and Micah Goldblum. LiveBench: A Challenging, Contamination-Free LLM Benchmark. arXiv p… find
  47. Zihuiwen Ye, Fraser Greenlee-Scott, Max Bartolo, Phil Blunsom, Jon Ander Campos, and Matthias Gallé. Improving Reward Models with Synthetic Critiques. arXiv preprint arXiv:2405.20850, 2024. find
  48. Weizhe Yuan, Pengfei Liu, and Matthias Gallé. LLMCrit: Teaching Large Language Models to Use Criteria. arXiv preprint arXiv:2403.01069, 2024. find
  49. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric P. Xing, Hao Zhang, Joseph E. Gonzalez, and Ion Stoica. Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena. In Alice Oh, Tristan Naumann, Amir Globerson, Kate Saenko, Morit…
  50. Jeffrey Zhou, Tianjian Lu, Swaroop Mishra, Siddhartha Brahma, Sujoy Basu, Yi Luan, Denny Zhou, and Le Hou. Instruction-Following Evaluation for Large Language Models. arXiv preprint arXiv:2311.07911, 2023.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/cook-2024-ticking-all-boxes-generated-checklists/