6 papers
Tell Model Where to Look: Mitigating Hallucinations in MLLMs by Vision-Guided Attention
Jianfei Zhao, Feng Zhang, Xin Sun +2
Visual attention serves as the primary mechanism through which MLLMs interpret visual information; however, its limited localization capability often leads to hallucinations. We ob…
ADHint: Adaptive Hints with Difficulty Priors for Reinforcement Learning
Feng Zhang, Zezhong Tan, Xinhong Ma +5
To address the limited capability expansion and low sample efficiency of Reinforcement Learning (RL), recent methods have integrated ''hints'' into post-training, which are prefix…
BagelVLA: Enhancing Long-Horizon Manipulation via Interleaved Vision-Language-Action Generation
Yucheng Hu, Jianke Zhang, Yuanfei Luo +9
Equipping embodied agents with the ability to reason about tasks, foresee physical outcomes, and generate precise actions is essential for general-purpose manipulation. While recen…
Cross-Layer Vision Smoothing: Enhancing Visual Understanding via Sustained Focus on Key Objects in Large Vision-Language Models
Jianfei Zhao, Feng Zhang, Xin Sun +2
Large Vision-Language Models (LVLMs) can accurately locate key objects in images, yet their attention to these objects tends to be very brief. Motivated by the hypothesis that sust…
Mitigating Hallucination in Large Vision-Language Models through Aligning Attention Distribution to Information Flow
Jianfei Zhao, Feng Zhang, Xin Sun +1
Due to the unidirectional masking mechanism, Decoder-Only models propagate information from left to right. LVLMs (Large Vision-Language Models) follow the same architecture, with v…
Cross-Image Contrastive Decoding: Precise, Lossless Suppression of Language Priors in Large Vision-Language Models
Jianfei Zhao, Feng Zhang, Xin Sun +3
Over-reliance on language priors is a major cause of hallucinations in Large Vision-Language Models (LVLMs), often leading to outputs that are linguistically plausible but visually…