6 papers
OneVision-Encoder: Codec-Aligned Sparsity as a Foundational Principle for Multimodal Intelligence
Feilong Tang, Xiang An, Yunyao Yan +16
Hypothesis. Artificial general intelligence is, at its core, a compression problem. Effective compression demands resonance: deep learning scales best when its architecture aligns…
LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal Training
Xiang An, Yin Xie, Kaicheng Yang +20
We present LLaVA-OneVision-1.5, a novel family of Large Multimodal Models (LMMs) that achieve state-of-the-art performance with significantly reduced computational and financial co…
PaCo-FR: Patch-Pixel Aligned End-to-End Codebook Learning for Facial Representation Pre-training
Yin Xie, Zhichao Chen, Zeyu Xiao +7
Facial representation pre-training is crucial for tasks like facial recognition, expression analysis, and virtual reality. However, existing methods face three key challenges: (1)…
Region-based Cluster Discrimination for Visual Representation Learning
Yin Xie, Kaicheng Yang, Xiang An +9
Learning visual representations is foundational for a broad spectrum of downstream tasks. Although recent vision-language contrastive models, such as CLIP and SigLIP, have achieved…
RealSyn: An Effective and Scalable Multimodal Interleaved Document Transformation Paradigm
Tiancheng Gu, Kaicheng Yang, Chaoyi Zhang +6
After pre-training on extensive image-text pairs, Contrastive Language-Image Pre-training (CLIP) demonstrates promising performance on a wide variety of benchmarks. However, a subs…
ViCToR: Improving Visual Comprehension via Token Reconstruction for Pretraining LMMs
Yin Xie, Kaicheng Yang, Peirou Liang +7
Large Multimodal Models (LMMs) often face a modality representation gap during pretraining: while language embeddings remain stable, visual representations are highly sensitive to…