Generator-Assistant Stepwise Rollback Framework for Large Language Model Agent

ACL ARR 2025 May Submission3264 Authors

Abstract:

Large language model (LLM) agents typically adopt a step-by-step reasoning framework, in which they interleave the processes of thinking and acting to accomplish the given task.
However, this paradigm faces a deep-rooted one-pass issue whereby each generated intermediate thought is plugged into the trajectory regardless of its correctness, which can cause irreversible error propagation.
To address the issue, this paper proposes a novel framework called Generator-Assistant Stepwise Rollback (GA-Rollback) to induce better decision-making for LLM agents.
Particularly, GA-Rollback utilizes a generator to interact with the environment and an assistant to examine each action produced by the generator, where the assistant triggers a rollback operation upon detection of incorrect actions.
Moreover, we introduce two additional strategies tailored for the rollback scenario to further improve its effectiveness.
Extensive experiments show that GA-Rollback achieves significant improvements over several strong baselines on three widely used benchmarks.
Our analysis further reveals that GA-Rollback can function as a robust plug-and-play module, integrating seamlessly with other methods.

Paper Type: Long

Research Area: Dialogue and Interactive Systems

Research Area Keywords: task-oriented, embodied agents, applications

Contribution Types: NLP engineering experiment, Data analysis

Languages Studied: English

Keywords: task-oriented, embodied agents, applications

Submission Number: 3264