LLMs that Reason and Act

🟦 This article is rated medium

Reading Time: 1 minute

Last updated on August 7, 2024

Takeaways

ReAct Extension: ReAct Systems enhance MRKL frameworks by combining reasoning with actions, enabling LLMs to improve complex task performance through iterative thought-action loops.

What is ReAct?

ReAct (Reason + Act) is a paradigm that enables Large Language Models (LLMs) to solve complex tasks through natural language reasoning and actions. It allows an LLM to perform certain actions, such as retrieving external information, and then reason based on the retrieved data.

ReAct systems extend Modular Reasoning, Knowledge, and Language (MRKL) systems by adding the ability to reason about the actions they can perform.

Example

Below is an example from HotPotQA, a question-answering dataset requiring complex reasoning. ReAct allows the LLM to reason about the question (Thought 1), take actions (e.g., querying Google) (Act 1). It then receives an observation (Obs 1) and continues the thought-action loop until reaching a conclusion (Act 3).

ReAct System (Yao et al.)

Readers with knowledge of Reinforcement Learning (RL) may recognize this process as similar to the classic RL loop of state, action and reward. ReAct provides some formalization for this in their paper.

Results

Google experimented with the PaLM LLM using ReAct, and the results showed promising improvements in complex reasoning tasks. ReAct was tested on datasets such as FEVER, focusing on fact extraction and verification.

ReAct Results (Yao et al.)

Sander Schulhoff

Sander Schulhoff is the Founder of Learn Prompting and an ML Researcher at the University of Maryland. He created the first open-source Prompt Engineering guide, reaching 3M+ people and teaching them to use tools like ChatGPT. Sander also led a team behind Prompt Report, the most comprehensive study of prompting ever done, co-authored with researchers from the University of Maryland, OpenAI, Microsoft, Google, Princeton, Stanford, and other leading institutions. This 76-page survey analyzed 1,500+ academic papers and covered 200+ prompting techniques.

Footnotes

Yao, S., Zhao, J., Yu, D., Du, N., Shafran, I., Narasimhan, K., & Cao, Y. (2022). ↩
Yang, Z., Qi, P., Zhang, S., Bengio, Y., Cohen, W. W., Salakhutdinov, R., & Manning, C. D. (2018). HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering. ↩
Chowdhery, A., Narang, S., Devlin, J., Bosma, M., Mishra, G., Roberts, A., Barham, P., Chung, H. W., Sutton, C., Gehrmann, S., Schuh, P., Shi, K., Tsvyashchenko, S., Maynez, J., Rao, A., Barnes, P., Tay, Y., Shazeer, N., Prabhakaran, V., … Fiedel, N. (2022). PaLM: Scaling Language Modeling with Pathways. ↩
Thorne, J., Vlachos, A., Christodoulopoulos, C., & Mittal, A. (2018). FEVER: a large-scale dataset for Fact Extraction and VERification. ↩