jisheng yin — accepted AI-conference papers

2025

Open Vision Reasoner: Transferring Linguistic Cognitive Behavior for Visual Reasoning

Yana Wei, Liang Zhao, Jianjian Sun, Kangheng Lin, jisheng yin, Jingcheng Hu +10

NeurIPS 2025poster

The remarkable reasoning capability of large language models (LLMs) stems from cognitive behaviors that emerge through reinforcement with verifiable rewards. This work investigates how to transfer this principle to Multimodal LLMs (MLLMs) to unlock advanced visual reasoning. We introduce a two-stage…

Cited by 0 Source ↗Scholar ↗

2025

Perception-R1: Pioneering Perception Policy with Reinforcement Learning

En Yu, Kangheng Lin, Liang Zhao, jisheng yin, Yana Wei, Yuang Peng +8

NeurIPS 2025poster

Inspired by the success of DeepSeek-R1, we explore the potential of rule-based reinforcement learning (RL) in MLLM post-training for perception policy learning. While promising, our initial experiments reveal that incorporating a thinking process through RL does not consistently lead to performance…

Cited by 0 Source ↗code ↗Scholar ↗