17 papers
Continual Video-MLLM Adaptation over Evolving Domains
Rui Cheng, Meixing Shi, Yuxiang Cai +3
Video multimodal large language models have shown strong capability in video understanding, yet their adaptation to sequentially evolving domains remains underexplored. In real-wor…
Spectral Evolution-Guided Token Pruning in Multimodal Large Language Models
Bin Chen, Yuxiang Cai, Yadan Luo +3
Reducing visual token redundancy is critical for accelerating Multimodal Large Language Models (MLLMs) without degrading cross-modal reasoning performance. Existing token pruning m…
Accelerating Multimodal Large Language Models with Prior-Corrected Token Reduction
Zengjie Chen, Yuxiang Cai, Jingcai Guo +3
Visual token reduction has emerged as an effective strategy for accelerating Multimodal Large Language Models (MLLMs). Many existing methods prune tokens by ranking text-visual att…
REKEY: Metadata-Grounded Visual-Key Regeneration for Contamination-Resilient VQA Evaluation
Tengjie Lin, Yutao Sun, Jingwei Ni +7
Static visual question answering (VQA) benchmarks age quickly: Once the items leak into training corpora, scores can reflect memorization rather than genuine visual ability, thus o…
VITAL: Visual-Semantic Dual Supervision for Enhanced and Interpretable Latent Reasoning in Medical MLLMs
Qiaoru Li, Shaotian Liang, Jintao Chen +4
Latent reasoning enables reasoning over continuous hidden states rather than explicit tokens, avoiding the language bottleneck and inference overhead of chain-of-thought for medica…
Which Pretraining Paradigm Better Serves Spatial Intelligence? An Empirical Comparison of Vision-Language and Video Generation Models
Haozhan Shen, Tiancheng Zhao, Kangjia Zhao +1
Spatial intelligence requires visual representations that capture both semantic objects and geometric structure in the physical world. To support this, two major pre-training schem…