Jianyi Zhang

DPad: Efficient Diffusion Language Models with Suffix Dropout

Accelerating Denoising Generative Models is as Easy as Predicting Second-Order Difference

Angles Don’t Lie: Unlocking Training‑Efficient RL Through the Model’s Own Signals

Min-K%++: Improved Baseline for Pre-Training Data Detection from Large Language Models

Proactive Privacy Amnesia for Large Language Models: Safeguarding PII with Negligible Impact on Model Utility

KVCOMM: Online Cross-context KV-cache Communication for Efficient LLM-based Multi-agent Systems

Sufficient Context: A New Lens on Retrieval Augmented Generation Systems

CoreInfer: Accelerating Large Language Model Inference with Semantics-Inspired Adaptive Sparse Activation

CoreMatching: A Co-adaptive Sparse Inference Framework with Token and Neuron Pruning for Comprehensive Acceleration of Vision-Language Models

SADA: Stability-guided Adaptive Diffusion Acceleration