4 papers
NaViL: Rethinking Scaling Properties of Native Multimodal Large Language Models under Data Constraints
Changyao Tian, Hao Li, Gen Luo +11
Compositional training has been the de-facto paradigm in existing Multimodal Large Language Models (MLLMs), where pre-trained vision encoders are connected with pre-trained LLMs th…
Docopilot: Improving Multimodal Models for Document-Level Understanding
Yuchen Duan, Zhe Chen, Yusong Hu +9
Despite significant progress in multimodal large language models (MLLMs), their performance on complex, multi-page document comprehension remains inadequate, largely due to the lac…
Learning Adaptive and Temporally Causal Video Tokenization in a 1D Latent Space
Yan Li, Changyao Tian, Renqiu Xia +7
We propose AdapTok, an adaptive temporal causal video tokenizer that can flexibly allocate tokens for different frames based on video content. AdapTok is equipped with a block-wise…
SynerGen-VL: Towards Synergistic Image Understanding and Generation with Vision Experts and Token Folding
Hao Li, Changyao Tian, Jie Shao +8
The remarkable success of Large Language Models (LLMs) has extended to the multimodal domain, achieving outstanding performance in image understanding and generation. Recent effort…