Chuanyang Zheng

ATTS: Asynchronous Test-Time Scaling via Conformal Prediction

Answer-Consistent Chain-of-thought Reinforcement Learning For Multi-modal Large Langauge Models

Understanding the Mixture-of-Experts with Nadaraya-Watson Kernel

Elevating Large Language Models with Module-of-Thought

SAS: Simulated Attention Score

DAPE V2: Process Attention Score as Feature Map for Length Extrapolation

ParallelComp: Parallel Long-Context Compressor for Length Extrapolation

UNComp: Uncertainty-Aware Long-context Compressor for Efficient Large Language Model Inference

UncertaintyRAG: Span Uncertainty Enhanced Long-Context Modeling for Retrieval-Augmented Generation

Efficient Multi-modal Large Language Models via Visual Token Grouping