Agent Instructs Large Language Models to be General Zero-Shot Reasoners

  • Nicholas Crispino
  • , Kyle Montgomery
  • , Fankun Zeng
  • , Dawn Song
  • , Chenguang Wang

Research output: Contribution to journalConference articlepeer-review

1 Scopus citations

Abstract

We introduce a method to improve the zero-shot reasoning abilities of large language models on general language understanding tasks. Specifically, we build an autonomous agent to instruct the reasoning process of large language models. To enable this, our agent only needs to generate a single set of instructions for each task. These instructions turn out to be extremely effective for improving the reasoning process of different large language models across all task instances. We show this approach further unleashes the zero-shot reasoning abilities of large language models to more tasks. We study the performance of our method on a wide set of datasets spanning generation, classification, and reasoning. We show that our method generalizes to most tasks and obtains state-of-the-art zero-shot performance on 20 of the 29 datasets that we evaluate. For instance, our method boosts the performance of state-of-the-art large language models by a large margin, including Vicuna-13b, Llama-2-70b-chat, and GPT-3.5 Turbo. Compared to zero-shot chain of thought, our improvement in reasoning is striking. With our method, Llama-2-70b-chat outperforms zero-shot GPT-3.5 Turbo significantly.

Original languageEnglish
Pages (from-to)9458-9549
Number of pages92
JournalProceedings of Machine Learning Research
Volume235
StatePublished - 2024
Event41st International Conference on Machine Learning, ICML 2024 - Vienna, Austria
Duration: Jul 21 2024Jul 27 2024

Fingerprint

Dive into the research topics of 'Agent Instructs Large Language Models to be General Zero-Shot Reasoners'. Together they form a unique fingerprint.

Cite this