17 papers · 1 filter
Self-Prophetic Decoding to Unlock Visual Search in LVLMs
Zhendong He, Qiyuan Dai, Guanbin Li +2
Large Vision-Language Models (LVLMs) are rapidly evolving toward true multimodal reasoning, with visual search representing a concrete instantiation of the thinking-with-images par…
Vision Transformers Need More Than Registers
Cheng Shi, Yizhou Yu, Sibei Yang
Vision Transformers (ViTs), when pre-trained on large-scale data, provide general-purpose representations for diverse downstream tasks. However, artifacts in ViTs are widely observ…
Chart Deep Research in LVLMs via Parallel Relative Policy Optimization
Jiajin Tang, Gaoyang, Wenjie Wang +2
With the rapid advancement of data science, charts have evolved from simple numerical presentation tools to essential instruments for insight discovery and decision-making support.…
WeaveTime: Stream from Earlier Frames into Emergent Memory in VideoLLMs
Yulin Zhang, Cheng Shi, Sibei Yang
Recent advances in Multimodal Large Language Models have greatly improved visual understanding and reasoning, yet their quadratic attention and offline training protocols make them…
Intervene-All-Paths: Unified Mitigation of LVLM Hallucinations across Alignment Formats
Jiaye Qian, Ge Zheng, Yuchen Zhu +1
Despite their impressive performance across a wide range of tasks, Large Vision-Language Models (LVLMs) remain prone to hallucination. In this study, we propose a comprehensive int…
Augmenting Moment Retrieval: Zero-Dependency Two-Stage Learning
Zhengxuan Wei, Jiajin Tang, Sibei Yang
Existing Moment Retrieval methods face three critical bottlenecks: (1) data scarcity forces models into shallow keyword-feature associations; (2) boundary ambiguity in transition r…