Yijiang Li

CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards

Vision Language Models Cannot Reason About Physical Transformation

Can 3D Vision-Language Models Truly Understand Natural Language?

Polymorphic: Plug-and-Play Visual Token Compression for Scalable VLMs

Core Knowledge Deficits in Multi-Modal Language Models

From Pixels to Tokens: Byte-Pair Encoding on Quantized Visual Modalities

EgoPrivacy: What Your First-Person Camera Says About You?

CogDevelop2K: Reversed Cognitive Development in Multi-modal Large Language Models

ARVideo: Autoregressive Pretraining for Self-Supervised Video Representation Learning

Beyond Finite Data: Towards Data-free Out-of-distribution Generalization via Extrapolation

Pseudo-Probability Unlearning: Towards Efficient and Privacy-Preserving Machine Unlearning

FedDTPT: Federated Discrete and Transferable Prompt Tuning for Black-Box Large Language Models