![]() | Draft article not currently submitted for review.
This is a draft Articles for creation (AfC) submission. It is not currently pending review. While there are no deadlines, abandoned drafts may be deleted after six months. To edit the draft click on the "Edit" tab at the top of the window. To be accepted, a draft should:
It is strongly discouraged to write about yourself, your business or employer. If you do so, you must declare it. Where to get help
How to improve a draft
You can also browse Wikipedia:Featured articles and Wikipedia:Good articles to find examples of Wikipedia's best writing on topics similar to your proposed article. Improving your odds of a speedy review To improve your odds of a faster review, tag your draft with relevant WikiProject tags using the button below. This will let reviewers know a new draft has been submitted in their area of interest. For instance, if you wrote about a female astronomer, you would want to add the Biography, Astronomy, and Women scientists tags. Editor resources
Last edited by Whpq (talk | contribs) 23 days ago. (Update) |
This article needs additional citations for verification. (July 2024) |
ReAct (Reasoning+Acting) is a prompting technique for large language models (LLMs) that combines reasoning and acting to improve performance on complex tasks. It was introduced by Yao et al. in 2022 in their paper "ReAct: Synergizing Reasoning and Acting in Language Models"[1].
Overview
editReAct prompts LLMs to generate both reasoning traces and task-specific actions in an interleaved manner. This allows the model to:
- Induce, track, and update action plans
- Handle exceptions
- Gather and incorporate external information
The key idea is to leverage the synergies between reasoning and acting, similar to how humans approach complex tasks.
How it works
editA ReAct prompt typically consists of:
- A task description
- Few-shot examples showing the desired format
- Alternating "Thought", "Action", and "Observation" steps
For example: ``` Question: What is the elevation range for the area that the eastern sector of the Colorado orogeny extends into?
Thought 1: I need to search for information about the Colorado orogeny and its eastern sector.
Action 1: Search[Colorado orogeny]
Observation 1: The Colorado orogeny was an episode of mountain building in Colorado and surrounding areas.
Thought 2: I need more specific information about the eastern sector.
Action 2: Lookup[eastern sector]
Observation 2: The eastern sector extends into the High Plains and is called the Central Plains orogeny.
Thought 3: Now I need to find the elevation range of the High Plains.
Action 3: Search[High Plains elevation range]
Observation 3: The High Plains rise in elevation from around 1,800 to 7,000 ft (550 to 2,130 m).
Thought 4: I have found the answer.
Action 4: Finish[1,800 to 7,000 ft ```
Applications
editReAct has been shown to be effective in various tasks, including:
- Question answering (e.g., HotpotQA)
- Fact verification (e.g., FEVER dataset)
- Decision-making in complex environments (e.g., ALFWorld, WebShop)
Advantages
editCompared to other prompting techniques, ReAct offers several benefits:
- Improved performance on knowledge-intensive tasks
- Better handling of multi-step reasoning
- Increased interpretability of model decisions
- Reduced hallucination through grounding in external information
Limitations
editSome limitations of ReAct include:
- Increased prompt length due to the inclusion of reasoning steps
- Potential for error propagation if early reasoning steps are incorrect
- Dependence on the quality of external information sources
See also
editReferences
edit[1] Yao, S., Zhao, J., Yu, D., Du, N., Shafran, I., Narasimhan, K., & Cao, Y. (2022). ReAct: Synergizing Reasoning and Acting in Language Models. arXiv preprint arXiv:2210.03629.