M$^{2}$Chat: Empowering VLM for Multimodal LLM Interleaved Text-Image Generation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chi, Xiaowei, Qi, Junbo, Zhang, Rongyu, Zhang, Shanghang, Liu, Qifeng, Guo, Yike
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916884442513408
author Chi, Xiaowei
Qi, Junbo
Zhang, Rongyu
Zhang, Shanghang
Liu, Qifeng
Guo, Yike
author_facet Chi, Xiaowei
Qi, Junbo
Zhang, Rongyu
Zhang, Shanghang
Liu, Qifeng
Guo, Yike
contents While current LLM chatbots like GPT-4V bridge the gap between human instructions and visual representations to enable text-image generations, they still lack efficient alignment methods for high-fidelity performance on multiple downstream tasks. In this paper, we propose \textbf{$M^{2}Chat$}, a novel unified multimodal LLM framework for generating interleaved text-image conversation across various scenarios. Specifically, we propose an $M^{3}Adapter$ that efficiently integrates granular low-level visual information and high-level semantic features from multi-modality prompts. Upon the well-aligned fused feature, $M^{3}Adapter$ tailors a learnable gating strategy to balance the model creativity and consistency across various tasks adaptively. Moreover, to further enhance the effectiveness of $M^{3}Adapter$ while preserving the coherence of semantic context comprehension, we introduce a two-stage $M^{3}FT$ fine-tuning strategy. This strategy optimizes disjoint groups of parameters for image-text alignment and visual-instruction respectively. Extensive experiments demonstrate our $M^{2}Chat$ surpasses state-of-the-art counterparts across diverse benchmarks, showcasing its prowess in interleaving generation, storytelling, and multimodal dialogue systems. The demo and code are available at \red{https://mattie-e.github.io/M2Chat.github.io}.
format Preprint
id arxiv_https___arxiv_org_abs_2311_17963
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle M$^{2}$Chat: Empowering VLM for Multimodal LLM Interleaved Text-Image Generation
Chi, Xiaowei
Qi, Junbo
Zhang, Rongyu
Zhang, Shanghang
Liu, Qifeng
Guo, Yike
Computer Vision and Pattern Recognition
While current LLM chatbots like GPT-4V bridge the gap between human instructions and visual representations to enable text-image generations, they still lack efficient alignment methods for high-fidelity performance on multiple downstream tasks. In this paper, we propose \textbf{$M^{2}Chat$}, a novel unified multimodal LLM framework for generating interleaved text-image conversation across various scenarios. Specifically, we propose an $M^{3}Adapter$ that efficiently integrates granular low-level visual information and high-level semantic features from multi-modality prompts. Upon the well-aligned fused feature, $M^{3}Adapter$ tailors a learnable gating strategy to balance the model creativity and consistency across various tasks adaptively. Moreover, to further enhance the effectiveness of $M^{3}Adapter$ while preserving the coherence of semantic context comprehension, we introduce a two-stage $M^{3}FT$ fine-tuning strategy. This strategy optimizes disjoint groups of parameters for image-text alignment and visual-instruction respectively. Extensive experiments demonstrate our $M^{2}Chat$ surpasses state-of-the-art counterparts across diverse benchmarks, showcasing its prowess in interleaving generation, storytelling, and multimodal dialogue systems. The demo and code are available at \red{https://mattie-e.github.io/M2Chat.github.io}.
title M$^{2}$Chat: Empowering VLM for Multimodal LLM Interleaved Text-Image Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.17963