Large Multimodal Models (LMMs) often face a modality representation gap during pretraining: while language embeddings remain stable, visual representations are highly sensitive to contextual noise (e.g., background clutter). To address this issue, we introduce a visual comprehension stage, which we call ViCToR (Visual Comprehension via Token Reconstruction), a novel pretraining framework for LMMs. ViCToR employs a learnable visual token pool and utilizes the Hungarian matching algorithm to select semantically relevant tokens from this pool for visual token replacement. Furthermore, by integrating a visual token reconstruction loss with dense semantic supervision, ViCToR can learn tokens which retain high visual detail, thereby enhancing the large language model's (LLM's) understanding of visual information. After pretraining on 3 million publicly accessible images and captions, ViCToR achieves state-of-the-art results, improving over LLaVA-NeXT-8B by 10.4%, 3.2%, and 7.2% on the MMStar, SEEDI, and RealWorldQA benchmarks, respectively. Code is available at https://github.com/deepglint/Victor.
@article{arxiv.2410.14332,
title = {ViCToR: Improving Visual Comprehension via Token Reconstruction for Pretraining LMMs},
author = {Yin Xie and Kaicheng Yang and Peirou Liang and Xiang An and Yongle Zhao and Yumeng Wang and Ziyong Feng and Roy Miles and Ismail Elezi and Jiankang Deng},
journal= {arXiv preprint arXiv:2410.14332},
year = {2025}
}