Shiyu Chang

7

Papers

40

Total Citations

Papers (7)

KVLink: Accelerating Large Language Models via Efficient KV Cache Reuse

NeurIPS 2025arXiv

Correcting Diffusion Generation through Resampling

Fictitious Synthetic Data Can Improve LLM Factuality via Prerequisite Learning

VSP: Diagnosing the Dual Challenges of Perception and Reasoning in Spatial Planning Tasks for MLLMs

Sparse Cocktail: Every Sparse Pattern Every Sparse Ratio All At Once

Speech Self-Supervised Learning Using Diffusion Model Synthetic Data

Decomposing Uncertainty for Large Language Models through Input Clarification Ensembling