English

Auto-Encoding Morph-Tokens for Multimodal LLM

Computer Vision and Pattern Recognition 2024-05-06 v1

Abstract

For multimodal LLMs, the synergy of visual comprehension (textual output) and generation (visual output) presents an ongoing challenge. This is due to a conflicting objective: for comprehension, an MLLM needs to abstract the visuals; for generation, it needs to preserve the visuals as much as possible. Thus, the objective is a dilemma for visual-tokens. To resolve the conflict, we propose encoding images into morph-tokens to serve a dual purpose: for comprehension, they act as visual prompts instructing MLLM to generate texts; for generation, they take on a different, non-conflicting role as complete visual-tokens for image reconstruction, where the missing visual cues are recovered by the MLLM. Extensive experiments show that morph-tokens can achieve a new SOTA for multimodal comprehension and generation simultaneously. Our project is available at https://github.com/DCDmllm/MorphTokens.

Keywords

Cite

@article{arxiv.2405.01926,
  title  = {Auto-Encoding Morph-Tokens for Multimodal LLM},
  author = {Kaihang Pan and Siliang Tang and Juncheng Li and Zhaoyu Fan and Wei Chow and Shuicheng Yan and Tat-Seng Chua and Yueting Zhuang and Hanwang Zhang},
  journal= {arXiv preprint arXiv:2405.01926},
  year   = {2024}
}

Comments

Accepted by ICML 2024

R2 v1 2026-06-28T16:15:15.355Z