Octopus: Embodied Vision-Language Programmer from Environmental Feedback

Jingkang Yang, Yuhao Dong, Shuai Liu, Bo Li, Ziyue Wang, Haoran Tan, Chencheng Jiang, Jiamu Kang, Yuanhan Zhang, Kaiyang Zhou, Ziwei Liu*

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

Abstract

Large vision-language models (VLMs) have achieved substantial progress in multimodal perception and reasoning. When integrated into an embodied agent, existing embodied VLM works either output detailed action sequences at the manipulation level or only provide plans at an abstract level, leaving a gap between high-level planning and real-world manipulation. To bridge this gap, we introduce Octopus, an embodied vision-language programmer that uses executable code generation as a medium to connect planning and manipulation. Octopus is designed to 1) proficiently comprehend an agent’s visual and textual task objectives, 2) formulate intricate action sequences, and 3) generate executable code. To facilitate Octopus model development, we introduce OctoVerse: a suite of environments tailored for benchmarking vision-based code generators on a wide spectrum of tasks, ranging from mundane daily chores in simulators to sophisticated interactions in complex video games such as Grand Theft Auto (GTA) and Minecraft. To train Octopus, we leverage GPT-4 to control an explorative agent that generates training data, i.e., action blueprints and corresponding executable code. We also collect feedback that enables an enhanced training scheme called Reinforcement Learning with Environmental Feedback (RLEF). Through a series of experiments, we demonstrate Octopus’s functionality and present compelling results, showing that the proposed RLEF refines the agent’s decision-making. By open-sourcing our simulation environments, dataset, and model architecture, we aspire to ignite further innovation and foster collaborative applications within the broader embodied AI community. The project page is available at https://choiszt.github.io/Octopus/.

Original languageEnglish
Title of host publicationComputer Vision – ECCV 2024 - 18th European Conference, Proceedings
EditorsAleš Leonardis, Elisa Ricci, Stefan Roth, Olga Russakovsky, Torsten Sattler, Gül Varol
PublisherSpringer Cham
Pages20-38
Number of pages19
ISBN (Electronic)9783031732324
ISBN (Print)9783031732317
DOIs
Publication statusE-pub ahead of print - 30 Sept 2024
Event18th European Conference on Computer Vision, ECCV 2024 - Milan, Italy
Duration: 29 Sept 20244 Oct 2024
https://eccv.ecva.net/Conferences/2024
https://link.springer.com/book/10.1007/978-3-031-73232-4

Publication series

NameLecture Notes in Computer Science
Volume15059
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference18th European Conference on Computer Vision, ECCV 2024
Country/TerritoryItaly
CityMilan
Period29/09/244/10/24
Internet address

Scopus Subject Areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this