This paper presents LLaVA-Plus (Large Language and Vision Assistants that Plug and Learn to Use Skills), a general-purpose multimodal assistant trained using an end-to-end approach that systematically expands the capabilities of large multimodal models (LMMs). LLaVA-Plus maintains a skill repository that contains a wide range of vision and vision-language pre-trained models (tools), and is able to activate relevant tools, given users’ multimodal inputs, to compose their execution results on the fly to fulfill many real-world tasks. To acquire the ability of using tools, LLaVA-Plus is trained on multimodal instruction-following data that we have curated. The training data covers many tool use examples of visual understanding, generation, external knowledge retrieval and their compositions. Empirical results show that LLaVA-Plus outperforms LLaVA in existing capabilities, and exhibits many new capabilities. Compared with tool-augmented LLMs, LLaVA-Plus is distinct in that the image query is directly grounded in and actively engaged throughout the entire human-AI interaction sessions, significantly improving tool use performance and enabling new scenarios.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

LLaVA-Plus: Learning to Use Tools for Creating Multimodal Agents

  • Shilong Liu,
  • Hao Cheng,
  • Haotian Liu,
  • Hao Zhang,
  • Feng Li,
  • Tianhe Ren,
  • Xueyan Zou,
  • Jianwei Yang,
  • Hang Su,
  • Jun Zhu,
  • Lei Zhang,
  • Jianfeng Gao,
  • Chunyuan Li

摘要

This paper presents LLaVA-Plus (Large Language and Vision Assistants that Plug and Learn to Use Skills), a general-purpose multimodal assistant trained using an end-to-end approach that systematically expands the capabilities of large multimodal models (LMMs). LLaVA-Plus maintains a skill repository that contains a wide range of vision and vision-language pre-trained models (tools), and is able to activate relevant tools, given users’ multimodal inputs, to compose their execution results on the fly to fulfill many real-world tasks. To acquire the ability of using tools, LLaVA-Plus is trained on multimodal instruction-following data that we have curated. The training data covers many tool use examples of visual understanding, generation, external knowledge retrieval and their compositions. Empirical results show that LLaVA-Plus outperforms LLaVA in existing capabilities, and exhibits many new capabilities. Compared with tool-augmented LLMs, LLaVA-Plus is distinct in that the image query is directly grounded in and actively engaged throughout the entire human-AI interaction sessions, significantly improving tool use performance and enabling new scenarios.