Embodied Multi-Modal Agent trained by an LLM from a Parallel TextWorld

Yijun Yang1,2,3; Tianyi Zhou4; Kanxue Li5; Dapeng Tao5; Lusong Li3; Li Shen3; Xiaodong He3; Jing Jiang2; Yuhui Shi1

10.1109/CVPR52733.2024.02482

2024-06-22

1063-6919

979-8-3503-5301-3

2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

16-22 June 2024

Seattle, WA, USA

While large language models (LLMs) excel in a simulated world of texts, they struggle to interact with the more realistic world without perceptions of other modalities such as visual or audio signals. Although vision-language models (VLMs) integrate LLM modules (1) aligned with static image features, and (2) may possess prior knowledge of world dynamics (as demonstrated in the text world), they have not been trained in an embodied visual world and thus cannot align with its dynamics. On the other hand, training an embodied agent in a noisy visual world without expert guidance is often chal-lenging and inefficient. In this paper, we train a VLM agent living in a visual world using an LLM agent excelling in a parallel text world. Specifically, we distill LLM's reflection outcomes (improved actions by analyzing mistakes) in a text world's tasks to finetune the VLM on the same tasks of the visual world, resulting in an Embodied Multi-Modal Agent (EMMA) quickly adapting to the visual world dy-namics. Such cross-modality imitation learning between the two parallel worlds is achieved by a novel DAgger-DPO algorithm, enabling EMMA to generalize to a broad scope of new tasks without any further guidance from the LLM expert. Extensive evaluations on the ALFWorld benchmark's diverse tasks highlight EMMA's superior performance to SOTA VLM-based agents, e.g., 20%-70% improvement in the success rate.

第一

1.Southern University of Science and Technology
2.University of Technology Sydney
3.JD Explore Academy
4.University of Maryland, College Park
5.Yunnan University

Yijun Yang,Tianyi Zhou,Kanxue Li,et al. Embodied Multi-Modal Agent trained by an LLM from a Parallel TextWorld[C],2024.