TRAD: Enhancing LLM Agents with Step-Wise Thought Retrieval and Aligned Decision
This work addresses a key bottleneck in LLM agent performance for tasks like web navigation and online shopping, offering a novel method to improve efficiency and generalization, though it is incremental relative to existing trajectory-based approaches.
The paper tackles the problem of selecting and effectively utilizing in-context examples for LLM agents in sequential decision-making tasks by proposing TRAD, a framework that uses step-wise thought retrieval and aligned decision to reduce noise and improve generalization, resulting in outperforming state-of-the-art models on benchmarks like ALFWorld and Mind2Web and enhancing success rates in real-world robotic process automation.
Numerous large language model (LLM) agents have been built for different tasks like web navigation and online shopping due to LLM's wide knowledge and text-understanding ability. Among these works, many of them utilize in-context examples to achieve generalization without the need for fine-tuning, while few of them have considered the problem of how to select and effectively utilize these examples. Recently, methods based on trajectory-level retrieval with task meta-data and using trajectories as in-context examples have been proposed to improve the agent's overall performance in some sequential decision making tasks. However, these methods can be problematic due to plausible examples retrieved without task-specific state transition dynamics and long input with plenty of irrelevant context. In this paper, we propose a novel framework (TRAD) to address these issues. TRAD first conducts Thought Retrieval, achieving step-level demonstration selection via thought matching, leading to more helpful demonstrations and less irrelevant input noise. Then, TRAD introduces Aligned Decision, complementing retrieved demonstration steps with their previous or subsequent steps, which enables tolerance for imperfect thought and provides a choice for balance between more context and less noise. Extensive experiments on ALFWorld and Mind2Web benchmarks show that TRAD not only outperforms state-of-the-art models but also effectively helps in reducing noise and promoting generalization. Furthermore, TRAD has been deployed in real-world scenarios of a global business insurance company and improves the success rate of robotic process automation.