MLLM as Retriever: Interactively Learning Multimodal Retrieval for Embodied Agents

TL;DR

This paper introduces MART, a novel method that fine-tunes multimodal large language models as retrievers for embodied agents, improving trajectory selection and task success in unseen environments through interaction-based preference learning and trajectory abstraction.

Contribution

It proposes a new paradigm of using fine-tuned MLLMs as retrievers for embodied agents, incorporating interaction data and trajectory abstraction to enhance multimodal retrieval.

Findings

Significant improvement in task success rates in unseen scenes.

Effective trajectory summarization with Trajectory Abstraction.

Enhanced understanding of trajectory milestones for better decision-making.

Abstract

MLLM agents demonstrate potential for complex embodied tasks by retrieving multimodal task-relevant trajectory data. However, current retrieval methods primarily focus on surface-level similarities of textual or visual cues in trajectories, neglecting their effectiveness for the specific task at hand. To address this issue, we propose a novel method, MLLM As ReTriever (MART), which enhances the performance of embodied agents by utilizing interaction data to fine-tune an MLLM retriever based on preference learning, such that the retriever fully considers the effectiveness of trajectories and prioritizes them for unseen tasks. We also introduce Trajectory Abstraction, a mechanism that leverages MLLMs' summarization capabilities to represent trajectories with fewer tokens while preserving key information, enabling agents to better comprehend milestones in the trajectory. Experimental results across various environments demonstrate our method significantly improves task success rates in unseen scenes compared to baseline methods. This work presents a new paradigm for multimodal retrieval in embodied agents, by fine-tuning a general-purpose MLLM as the retriever to assess trajectory effectiveness. All the code for benchmark tasks, simulator modifications, and the MLLM retriever is available at https://github.com/PKU-RL/MART.