4 papers
RLVMR: Reinforcement Learning with Verifiable Meta-Reasoning Rewards for Robust Long-Horizon Agents
Zijing Zhang, Ziyang Chen, Mingxiao Li +2
The development of autonomous agents for complex, long-horizon tasks is a central goal in AI. However, dominant training paradigms face a critical limitation: reinforcement learnin…
VISTA: Enhancing Vision-Text Alignment in MLLMs via Cross-Modal Mutual Information Maximization
Mingxiao Li, Na Su, Fang Qu +5
Current multimodal large language models (MLLMs) face a critical challenge in modality alignment, often exhibiting a bias towards textual information at the expense of other modali…
From Visuals to Vocabulary: Establishing Equivalence Between Image and Text Token Through Autoregressive Pre-training in MLLMs
Mingxiao Li, Fang Qu, Zhanpeng Chen +5
While MLLMs perform well on perceptual tasks, they lack precise multimodal alignment, limiting performance. To address this challenge, we propose Vision Dynamic Embedding-Guided Pr…
Advancing General Multimodal Capability of Vision-language Models with Pyramid-descent Visual Position Encoding
Zhanpeng Chen, Mingxiao Li, Ziyang Chen +3
Vision-language Models (VLMs) have shown remarkable capabilities in advancing general artificial intelligence, yet the irrational encoding of visual positions persists in inhibitin…