Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2405.16334 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929393049272320 |
|---|---|
| author | Wang, Haoyu Li, Tao Deng, Zhiwei Roth, Dan Li, Yang |
| author_facet | Wang, Haoyu Li, Tao Deng, Zhiwei Roth, Dan Li, Yang |
| contents | In this work, we introduce a novel approach that equips LLM agents with introspection, enhancing consistency and adaptability in solving complex tasks. Our approach prompts LLM agents to decompose a given task into manageable subtasks (i.e., to make a plan), and to continuously introspect upon the suitability and results of their actions. %; and when necessary, to explore ``the road not taken.'' We implement a three-fold introspective intervention: 1) anticipatory reflection on potential failures and alternative remedy before action execution, 2) post-action alignment with subtask objectives and backtracking with remedy to ensure utmost effort in plan execution, and 3) comprehensive review upon plan completion for future strategy refinement. By deploying and experimenting with this methodology -- a zero-shot approach -- within WebArena for practical tasks in web environments, our agent demonstrates superior performance with a success rate of 23.5% over existing zero-shot methods by 3.5%. The experimental results suggest that our introspection-driven approach not only enhances the agent's ability to navigate unanticipated challenges through a robust mechanism of plan execution, but also improves efficiency by reducing the number of trials and plan revisions by 45% needed to achieve a task. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_16334 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Devil's Advocate: Anticipatory Reflection for LLM Agents Wang, Haoyu Li, Tao Deng, Zhiwei Roth, Dan Li, Yang Artificial Intelligence In this work, we introduce a novel approach that equips LLM agents with introspection, enhancing consistency and adaptability in solving complex tasks. Our approach prompts LLM agents to decompose a given task into manageable subtasks (i.e., to make a plan), and to continuously introspect upon the suitability and results of their actions. %; and when necessary, to explore ``the road not taken.'' We implement a three-fold introspective intervention: 1) anticipatory reflection on potential failures and alternative remedy before action execution, 2) post-action alignment with subtask objectives and backtracking with remedy to ensure utmost effort in plan execution, and 3) comprehensive review upon plan completion for future strategy refinement. By deploying and experimenting with this methodology -- a zero-shot approach -- within WebArena for practical tasks in web environments, our agent demonstrates superior performance with a success rate of 23.5% over existing zero-shot methods by 3.5%. The experimental results suggest that our introspection-driven approach not only enhances the agent's ability to navigate unanticipated challenges through a robust mechanism of plan execution, but also improves efficiency by reducing the number of trials and plan revisions by 45% needed to achieve a task. |
| title | Devil's Advocate: Anticipatory Reflection for LLM Agents |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2405.16334 |