Elamur Research⌕
☰
home / papers / 2023 / INSTRUCTEVAL: Towards Holistic Evaluation of Instruction-Tuned Large Language Models

INSTRUCTEVAL: Towards Holistic Evaluation of Instruction-Tuned Large Language Models

Paper · 2023 · Yew Ken Chia, Pengfei Hong, Lidong Bing et al.

Summary

Instruction-tuned LLMs such as GPT-4 are widely used, but their real abilities are hard to judge. Many are closed systems, and open-source releases often come with informal claims, such as matching most of ChatGPT's quality, that lack solid evidence. The authors propose INSTRUCTEVAL, an evaluation suite that tests three areas together: problem-solving, writing, and alignment with human values. Tasks with clear answers are scored automatically, while open-ended writing is rated qualitatively, for example on coherence and relevance. The authors also survey open-source foundation models, instruction datasets and instructed models. This lets them link performance to the pretraining base, the tuning data or the training method. According to their analysis, how good the instruction data is appears to matter more than anything else for improving performance. Open models write well but lag clearly in problem-solving and alignment. The work provides a reproducible benchmark and a public leaderboard, which help separate real progress from hype among fast-moving open-source LLMs.

Key points

  • Problem-solving is measured with MMLU, BBH, DROP, CRASS and HumanEval.
  • Writing covers informative, professional, argumentative and creative tasks. The data is released as the IMPACT dataset (declare-lab/InstructEvalImpact on Hugging Face).
  • Alignment is assessed on harmlessness, honesty and helpfulness.
  • Leaderboard: it compares over 60 open-source LLMs, and the paper analyses 10 of them in depth, chosen by architecture, instruction data and pretraining. Code is at github.com/declare-lab/instruct-eval.
  • Surveyed resources:
    • 10 foundation models, e.g. LLaMA trained on 1.4T tokens, plus T5, UL2, GLM, RWKV and Pythia.
    • 10 instruction datasets, e.g. Flan Collection (15M), Alpaca Data (52K), ShareGPT (70K) and Databricks Dolly (15K).
    • Instructed models such as Flan-T5, Vicuna, Alpaca and ChatGLM. ChatGLM is the only one listed with RLHF; all others use supervised tuning.

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

🎓Singapore University of Technology and DesignSingapore · Singapore33 works · 10,760 citations

📚 Shelves 5The field of science this work belongs to (OpenAlex fields).Computer Science · 9,086 worksArtificial Intelligence · 8,889 worksArts and Humanities · 9,296 worksLanguage and Linguistics · 9,206 worksComputer Vision and Pattern Recognition · 1,282 works

🧭 Topics 1Research subjects and languages the work is about.Language models & pretraining · 1,343 works

Cite

Cite as Yew Ken Chia, Pengfei Hong, Lidong Bing and Soujanya Poria. 2023. INSTRUCTEVAL: Towards Holistic Evaluation of Instruction-Tuned Large Language Models. arXiv:2306.04757.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2306.04757 · arXiv 2306.04757

@misc{chia2023instructeval,
  title = {{INSTRUCTEVAL: Towards Holistic Evaluation of Instruction-Tuned Large Language Models}},
  author = {Yew Ken Chia, Pengfei Hong, Lidong Bing and Soujanya Poria},
  year = {2023},
  doi = {10.48550/arxiv.2306.04757},
  eprint = {2306.04757},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2306.04757}
}

References

39 references, 24 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. OpenAI. Gpt-4 technical report. ArXiv, abs/2303.08774, 2023.
  2. Sid Black, Stella Biderman, Eric Hallahan, Quentin Anthony, Leo Gao, Laurence Golding, Horace He, Connor Leahy, Kyle McDonell, Jason Phang, Michael Pieler, USVSN Sai Prashanth, Shivanshu Purohit, Laria Reynolds, Jonathan Tow, Ben Wang, and Samuel Weinbach. Gpt-neox-20b: An open-source autoregressive…
  3. StabilityAI. Stablelm: Stability ai language models, April 2023. URL https://github.com/ Stability-AI/StableLM. ↗
  4. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aur’elien Rodriguez, Armand Joulin, Edouard Grave, and Guillaume Lample. Llama: Open and efficient foundation language models. ArXiv, abs/23…
  5. Stella Biderman, Hailey Schoelkopf, Quentin Anthony, Herbie Bradley, Kyle O’Brien, Eric Hallahan, Mohammad Aflah Khan, Shivanshu Purohit, USVSN Sai Prashanth, Edward Raff, Aviya Skowron, Lintang Sutawika, and Oskar van der Wal. Pythia: A suite for analyzing large language models across training and… find
  6. Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen, Christopher Dewan, Mona Diab, Xian Li, Xi Victoria Lin, Todor Mihaylov, Myle Ott, Sam Shleifer, Kurt Shuster, Daniel Simig, Punit Singh Koura, Anjali Sridhar, Tianlu Wang, and Luke Zettlemoyer. Opt: Open pre-trained tr…
  7. Yi Tay, Mostafa Dehghani, Vinh Q. Tran, Xavier Garcia, Jason Wei, Xuezhi Wang, Hyung Won Chung, Dara Bahri, Tal Schuster, Steven Zheng, Denny Zhou, Neil Houlsby, and Donald Metzler. UL2: Unifying language learning paradigms. In The Eleventh International Conference on Learning Representations, 2023.…
  8. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research, 21(140):1–67, 2020. URL http://jmlr.org/papers/v21/20… ↗
  9. Zhengxiao Du, Yujie Qian, Xiao Liu, Ming Ding, Jiezhong Qiu, Zhilin Yang, and Jie Tang. GLM: General language model pretraining with autoregressive blank infilling. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 320–335, Dubl…
  10. Bo Peng, Eric Alcaide, Quentin Anthony, Alon Albalak, Samuel Arcadinho, Huanqi Cao, Xin Cheng, Michael Chung, Matteo Grella, Kranthi Kiran GV, Xuzheng He, Haowen Hou, Przemyslaw Kazienko, Jan Kocon, Jiaming Kong, Bartlomiej Koptyra, Hayden Lau, Krishna Sri Ipsit Mantri, Ferdinand Mom, Atsushi Saito,… find
  11. MosaicML. Mpt-7b: A new standard for open-source, commercially usable llms, May 2023. URL https://www.mosaicml.com/blog/mpt-7b. ↗
  12. Rohan Taori, Ishaan Gulrajani, Tianyi Zhang, Yann Dubois, Xuechen Li, Carlos Guestrin, Percy Liang, and Tatsunori B. Hashimoto. Stanford alpaca: An instruction-following llama model, 2023. URL https://github.com/tatsu-lab/stanford_alpaca. ↗
  13. Shayne Longpre, Le Hou, Tu Vu, Albert Webson, Hyung Won Chung, Yi Tay, Denny Zhou, Quoc V Le, Barret Zoph, Jason Wei, et al. The flan collection: Designing data and methods for effective instruction tuning. arXiv preprint arXiv:2301.13688, 2023.
  14. 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, 2023. find
  15. Swaroop Mishra, Daniel Khashabi, Chitta Baral, and Hannaneh Hajishirzi. Cross-task generalization via natural language crowdsourcing instructions. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3470–3487, Dublin, Ireland, May… find
  16. Wei-Lin Chiang, Zhuohan Li, Zi Lin, Ying Sheng, Zhanghao Wu, Hao Zhang, Lianmin Zheng, Siyuan Zhuang, Yonghao Zhuang, Joseph E. Gonzalez, Ion Stoica, and Eric P. Xing. Vicuna: An open-source chatbot impressing gpt-4 with 90%* chatgpt quality, March 2023. URL https://vicuna.lmsys.org. ↗
  17. Victor Sanh, Albert Webson, Colin Raffel, Stephen Bach, Lintang Sutawika, Zaid Alyafeai, Antoine Chaffin, Arnaud Stiegler, Arun Raja, Manan Dey, M Saiful Bari, Canwen Xu, Urmish Thakker, Shanya Sharma Sharma, Eliza Szczechla, Taewoon Kim, Gunjan Chhablani, Nihal Nayak, Debajyoti Datta, Jonathan Chan…
  18. Databricks Labs. Dolly, 2023. URL https://github.com/databrickslabs/dolly. ↗
  19. Andreas Köpf, Yannic Kilcher, Dimitri von Rütte, Sotiris Anagnostidis, Zhi-Rui Tam, Keith Stevens, Abdullah Barhoum, Nguyen Minh Duc, Oliver Stanley, Richárd Nagyfi, Shahul ES, Sameer Suri, David Glushkov, Arnav Dantuluri, Andrew Maguire, Christoph Schuhmann, Huu Nguyen, and Alexander Mattick. Opena…
  20. 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
  21. LAION-AI. Open-Assistant. https://github.com/LAION-AI/Open-Assistant, 2023. ↗
  22. Srinivasan Iyer, Xi Victoria Lin, Ramakanth Pasunuru, Todor Mihaylov, Daniel Simig, Ping Yu, Kurt Shuster, Tianlu Wang, Qing Liu, Punit Singh Koura, Xian Li, Brian O’Horo, Gabriel Pereyra, Jeff Wang, Christopher Dewan, Asli Celikyilmaz, Luke Zettlemoyer, and Ves Stoyanov. Opt-iml: Scaling language m…
  23. Yizhong Wang, Swaroop Mishra, Pegah Alipoormolabashi, Yeganeh Kordi, Amirreza Mirzaei, Anjana Arunkumar, Arjun Ashok, Arut Selvan Dhanasekaran, Atharva Naik, David Stap, Eshaan Pathak, Giannis Karamanolakis, Haizhi Gary Lai, Ishan Purohit, Ishani Mondal, Jacob Anderson, Kirby Kuznia, Krima Doshi, Ma… find
  24. Yew Ken Chia, Pengfei Hong, and Soujanya Poria. Flan-alpaca: Instruction tuning from humans and machines, March 2023. URL https://github.com/declare-lab/flan-alpaca. ↗
  25. Hyung Won Chung, Le Hou, S. Longpre, Barret Zoph, Yi Tay, William Fedus, Eric Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Dasha Valter, Sharan Narang, Gaurav Mishra, Adams Wei Yu, Vincent Zhao, Ya…
  26. Aohan Zeng, Xiao Liu, Zhengxiao Du, Zihan Wang, Hanyu Lai, Ming Ding, Zhuoyi Yang, Yifan Xu, Wendi Zheng, Xiao Xia, Weng Lam Tam, Zixuan Ma, Yufei Xue, Jidong Zhai, Wenguang Chen, P. Zhang, Yuxiao Dong, and Jie Tang. Glm-130b: An open bilingual pre-trained model. ArXiv, abs/2210.02414, 2022.
  27. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. Measuring massive multitask language understanding. In International Conference on Learning Representations, 2021. URL https://openreview.net/forum?id=d7KBjmI3GmQ. ↗
  28. Wanjun Zhong, Ruixiang Cui, Yiduo Guo, Yaobo Liang, Shuai Lu, Yanlin Wang, Amin Saied, Weizhu Chen, and Nan Duan. Agieval: A human-centric benchmark for evaluating foundation models, 2023.
  29. Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao, Abu Awal Md Shoeb, Abubakar Abid, Adam Fisch, Adam R. Brown, Adam Santoro, Aditya Gupta, Adrià Garriga-Alonso, Agnieszka Kluska, Aitor Lewkowycz, Akshat Agarwal, Alethea Power, Alex Ray, Alex Warstadt, Alexander W. Kocurek, Ali Safaya, Ali Tazarv, Al…
  30. Mirac Suzgun, Nathan Scales, Nathanael Scharli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc V. Le, Ed Huai hsin Chi, Denny Zhou, and Jason Wei. Challenging big-bench tasks and whether chain-of-thought can solve them. ArXiv, abs/2210.09261, 2022.
  31. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde, Jared Kaplan, Harrison Edwards, Yura Burda, Nicholas Joseph, Greg Brockman, Alex Ray, Raul Puri, Gretchen Krueger, Michael Petrov, Heidy Khlaaf, Girish Sastry, Pamela Mishkin, Brooke Chan, Scott Gray, Nick Ryder, Mikhail Pavlov, Aleth…
  32. Amanda Askell, Yuntao Bai, Anna Chen, Dawn Drain, Deep Ganguli, Tom Henighan, Andy Jones, Nicholas Joseph, Ben Mann, Nova DasSarma, Nelson Elhage, Zac Hatfield-Dodds, Danny Hernandez, Jackson Kernion, Kamal Ndousse, Catherine Olsson, Dario Amodei, Tom Brown, Jack Clark, Sam McCandlish, Chris Olah, a… find
  33. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35: 27730–27744, 2022.
  34. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685, 2021.
  35. Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners. Advances in neural information processing systems, 33:1877–1901, 2020.
  36. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Ed Chi, Quoc Le, and Denny Zhou. Chain of thought prompting elicits reasoning in large language models. arXiv preprint arXiv:2201.11903, 2022.
  37. Zhiyong Wu, Yaoxiang Wang, Jiacheng Ye, and Lingpeng Kong. Self-adaptive in-context learning: An information compression perspective for in-context example selection and ordering, 2023. find
  38. Yao Lu, Max Bartolo, Alastair Moore, Sebastian Riedel, and Pontus Stenetorp. Fantastically ordered prompts and where to find them: Overcoming few-shot prompt order sensitivity. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8…
  39. Jiachang Liu, Dinghan Shen, Yizhe Zhang, Bill Dolan, Lawrence Carin, and Weizhu Chen. What makes good in-context examples for GPT-3? In Proceedings of Deep Learning Inside Out (DeeLIO 2022): The 3rd Workshop on Knowledge Extraction and Integration for Deep Learning Architectures, pages 100–114, Dubl…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/chia-2023-instructeval-towards-holistic-evaluation-instruction/