Reflection-Based Memory For Web navigation Agents
This addresses the issue of inefficiency and error repetition in web navigation agents, offering a domain-specific improvement.
The paper tackles the problem of web navigation agents repeating mistakes due to lack of memory by introducing Reflection-Augment Planning (ReAP), which uses self-reflections on past experiences to improve baseline results by 11 points overall and 29 points on previously failed tasks.
Web navigation agents have made significant progress, yet current systems operate with no memory of past experiences -- leading to repeated mistakes and an inability to learn from previous interactions. We introduce Reflection-Augment Planning (ReAP), a web navigation system to leverage both successful and failed past experiences using self-reflections. Our method improves baseline results by 11 points overall and 29 points on previously failed tasks. These findings demonstrate that reflections can transfer to different web navigation tasks.