PPL
Haoyu Wang
Papers (7)
BLINK: Multimodal Large Language Models Can See but Not PerceiveStructure-Aware RAG: Structured Retrieval Augmented Generation from Noisy Data for Conversational AgentsGAMMA: Global Bit Allocation for Mixed-Precision Models under Arbitrary BudgetsFormalize, Don't Optimize: The Heuristic Trap in LLM-Generated Combinatorial SolversTowards Understanding Continual Factual Knowledge Acquisition of Language Models: From Theory to AlgorithmOne Turn Too Late: Response-Aware Defense Against Hidden Malicious Intent in Multi-Turn DialogueTowards General Preference Alignment: Diffusion Models at Nash Equilibrium