Yiyu Shi

Post-Processing Approach for Distributive Fairness in Multi-Class Federated Learning

Driving Through Uncertainty: Risk-Averse Control with LLM Commonsense for Autonomous Driving under Perception Deficits

CompressKV: Semantic Retrieval Heads Know What Tokens are Not Important Before Generation

Large Language Models have Intrinsic Self-Correction Ability

One Training Fits All: Generalized Data Condensation via Mixture-of-Information Bottleneck Guidance

Empirical Guidelines for Deploying LLMs onto Resource-constrained Edge Devices