WebATLAS: An LLM Agent with Experience-Driven Memory and Action Simulation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911328778584064 |
|---|---|
| author | Cheng, Jiali Kumar, Anjishnu Lal, Roshan Rajasekaran, Rishi Ramezani, Hani Khan, Omar Zia Rokhlenko, Oleg Chiu-Webster, Sunny Hua, Gang Amiri, Hadi |
| author_facet | Cheng, Jiali Kumar, Anjishnu Lal, Roshan Rajasekaran, Rishi Ramezani, Hani Khan, Omar Zia Rokhlenko, Oleg Chiu-Webster, Sunny Hua, Gang Amiri, Hadi |
| contents | Large Language Model (LLM) web agents often struggle with long-horizon web navigation and web task completion in new websites, producing inefficient action sequences unless fine-tuned on environment-specific data. We show that experience-driven memory, combined with look-ahead action simulation, is sufficient for LLM agents to adapt to unseen web environments by remembering past failures and predicting the consequences of future actions. We introduce WebATLAS (Actor-Critic Task-completion with Look-ahead Action Simulation), a memory-augmented LLM web agent that learns a lightweight internal model of the environment from interaction experience and performs hypothetical action rollouts before acting in the real world. WebATLAS builds a persistent cognitive map via curiosity-driven exploration, stores interaction outcomes as experience-based memory, and evaluates candidate actions in cognitive space using a planner--simulator--critic loop. This enables the agent to reuse past experience, avoid previously unsuccessful behaviors, and generate more efficient plans. We evaluate WebATLAS on the WebArena-Lite benchmark for autonomous web navigation and demonstrate a success rate of 63%, outperforming the previous state-of-the-art at 53.9%. Unlike previous systems, our modular architecture requires no website-specific LLM fine-tuning. Ablation studies confirm that experience-driven memory, look-ahead action simulation, and hierarchical replanning play complementary roles in enabling robust, training-free web agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_22732 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WebATLAS: An LLM Agent with Experience-Driven Memory and Action Simulation Cheng, Jiali Kumar, Anjishnu Lal, Roshan Rajasekaran, Rishi Ramezani, Hani Khan, Omar Zia Rokhlenko, Oleg Chiu-Webster, Sunny Hua, Gang Amiri, Hadi Machine Learning Artificial Intelligence Computation and Language Information Retrieval Multiagent Systems Robotics Large Language Model (LLM) web agents often struggle with long-horizon web navigation and web task completion in new websites, producing inefficient action sequences unless fine-tuned on environment-specific data. We show that experience-driven memory, combined with look-ahead action simulation, is sufficient for LLM agents to adapt to unseen web environments by remembering past failures and predicting the consequences of future actions. We introduce WebATLAS (Actor-Critic Task-completion with Look-ahead Action Simulation), a memory-augmented LLM web agent that learns a lightweight internal model of the environment from interaction experience and performs hypothetical action rollouts before acting in the real world. WebATLAS builds a persistent cognitive map via curiosity-driven exploration, stores interaction outcomes as experience-based memory, and evaluates candidate actions in cognitive space using a planner--simulator--critic loop. This enables the agent to reuse past experience, avoid previously unsuccessful behaviors, and generate more efficient plans. We evaluate WebATLAS on the WebArena-Lite benchmark for autonomous web navigation and demonstrate a success rate of 63%, outperforming the previous state-of-the-art at 53.9%. Unlike previous systems, our modular architecture requires no website-specific LLM fine-tuning. Ablation studies confirm that experience-driven memory, look-ahead action simulation, and hierarchical replanning play complementary roles in enabling robust, training-free web agents. |
| title | WebATLAS: An LLM Agent with Experience-Driven Memory and Action Simulation |
| topic | Machine Learning Artificial Intelligence Computation and Language Information Retrieval Multiagent Systems Robotics |
| url | https://arxiv.org/abs/2510.22732 |