Wei Liu

FastEdit: Low-Rank Structured Regularization for Efficient Model Editing

Enhancing RAG with Active Learning on Conversation Records: Reject Incapables and Answer Capables

Model Editing is Over: Revealing Its Illusory Success and Fragile Foundation

Quantifying Distributional Invariance in Causal Subgraph for IRM-Free Graph Generalization

Breaking Free from MMI: A New Frontier in Rationalization by Probing Input Utilization

Adversarial Cooperative Rationalization: The Risk of Spurious Correlations in Even Clean Datasets

A Decoupled Learning Framework for Neural Marked Temporal Point Process

Attacking for Inspection and Instruction: Attack Techniques Can Aid In Interpretability