Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them

By Mirac Suzgun, Nathan Scales, Nathanael Scharli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc V. Le, Ed H. Chi, Denny Zhou and 1 more. 2022

2,044 citations recorded by the source corpus.

BIG-Bench (Srivastava et al., 2022) is a diverse evaluation suite that focuses on tasks believed to be beyond the capabilities of current language models. Language models have already made good progress on this benchmark, with the best model in the BIG-Bench paper outperforming average reported human-rater results on 65% of the BIG-Bench tasks via few-shot prompting. But on what tasks do language models fall short of average human-rater performance, and are those tasks actually unsolvable by current language models? In this work, we focus on a suite of 23 challenging BIG-Bench tasks which we call BIG-Bench Hard (BBH). These are the task for which prior language model evaluations did not outperform the average human-rater. We find that applying chain-of-thought (CoT) prompting to BBH tasks enables PaLM to surpass the average human-rater performance on 10 of the 23 tasks, and Codex (code-davinci-002) to surpass the average human-rater performance on 17 of the 23 tasks. Since many tasks in BBH require multi-step reasoning, few-shot prompting without CoT, as done in the BIG-Bench evaluations (Srivastava et al., 2022), substantially underestimates the best performance and capabilities of language models, which is better captured via CoT prompting. As further analysis, we explore the interaction between CoT and model scale on BBH, finding that CoT enables emergent task performance on several BBH tasks with otherwise flat scaling curves.

Subfields: Artificial Intelligence, Health Informatics, Prompting & Reasoning, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

  1. Mohit Bansal
  2. Luke S. Zettlemoyer
  3. Percy Liang
  4. Jianfeng Gao
  5. Graham Neubig
  6. Quoc V. Le
  7. Noah A. Smith
  8. Yejin Choi

In Health Informatics

  1. Noah A. Smith
  2. Percy Liang
  3. Melvin Johnson
  4. Oren Etzioni
  5. Dan Jurafsky
  6. Luciano Floridi
  7. Luke S. Zettlemoyer
  8. Xiao Liu

In Prompting & Reasoning

  1. Denny Zhou
  2. Luke S. Zettlemoyer
  3. Hannaneh Hajishirzi
  4. Wenhu Chen
  5. Tushar Khot
  6. Jie Huang
  7. Ashish Sabharwal
  8. Xinyun Chen

In Evaluation & Benchmarks

  1. Mohit Bansal
  2. Ashish Sabharwal
  3. Hannaneh Hajishirzi
  4. Luke S. Zettlemoyer
  5. Matt Gardner
  6. Yonatan Belinkov
  7. Yejin Choi
  8. Tushar Khot

Citations in the corpus