6 papers
Perceive, Interact, Reason: Building Tool-Augmented Visual Agents for Spatial Reasoning
Changye Li, Meng Lu, Yi Wu +1
While recent vision-language models (VLMs) demonstrate strong multimodal understanding, they remain limited in spatial reasoning tasks that require active evidence acquisition and…
EGM: Efficient Visual Grounding Language Models
Guanqi Zhan, Changye Li, Zhijian Liu +4
Visual grounding is an essential capability of Visual Language Models (VLMs) to understand the real physical world. Previous state-of-the-art grounding visual language models usual…
Learning Personalized Driving Styles via Reinforcement Learning from Human Feedback
Derun Li, Changye Li, Yue Wang +9
Generating human-like and adaptive trajectories is essential for autonomous driving in dynamic environments. While generative models have shown promise in synthesizing feasible tra…
Language Models Resist Alignment: Evidence From Data Compression
Jiaming Ji, Kaile Wang, Tianyi Qiu +7
Large language models (LLMs) may exhibit unintended or undesirable behaviors. Recent works have concentrated on aligning LLMs to mitigate harmful outputs. Despite these efforts, so…
SAE-V: Interpreting Multimodal Models for Enhanced Alignment
Hantao Lou, Changye Li, Jiaming Ji +1
With the integration of image modality, the semantic space of multimodal large language models (MLLMs) is more complex than text-only models, making their interpretability more cha…
Towards Efficient Collaboration via Graph Modeling in Reinforcement Learning
Wenzhe Fan, Zishun Yu, Chengdong Ma +3
In multi-agent reinforcement learning, a commonly considered paradigm is centralized training with decentralized execution. However, in this framework, decentralized execution rest…