Auto-Encoding Morph-Tokens for Multimodal LLM

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Pan, Kaihang, Tang, Siliang, Li, Juncheng, Fan, Zhaoyu, Chow, Wei, Yan, Shuicheng, Chua, Tat-Seng, Zhuang, Yueting, Zhang, Hanwang
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911864845238272
author Pan, Kaihang
Tang, Siliang
Li, Juncheng
Fan, Zhaoyu
Chow, Wei
Yan, Shuicheng
Chua, Tat-Seng
Zhuang, Yueting
Zhang, Hanwang
author_facet Pan, Kaihang
Tang, Siliang
Li, Juncheng
Fan, Zhaoyu
Chow, Wei
Yan, Shuicheng
Chua, Tat-Seng
Zhuang, Yueting
Zhang, Hanwang
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2405_01926
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Auto-Encoding Morph-Tokens for Multimodal LLM
Pan, Kaihang
Tang, Siliang
Li, Juncheng
Fan, Zhaoyu
Chow, Wei
Yan, Shuicheng
Chua, Tat-Seng
Zhuang, Yueting
Zhang, Hanwang
Computer Vision and Pattern Recognition
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.
title Auto-Encoding Morph-Tokens for Multimodal LLM
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.01926