Yida Wang

PLoRA: Efficient LoRA Hyperparameter Tuning for Large Language Models

TritonRL: Training LLMs to Think and Code Triton Without Cheating

EmbedMol: An Open Billion-scale Molecular Embedding Dataset for Molecular Discovery

Block-Diagonal LoRA for Eliminating Communication Overhead in Tensor Parallel LoRA Serving

KVFlow: Efficient Prefix Caching for Accelerating LLM-Based Multi-Agent Workflows