Jian Liang

Taming Momentum: Rethinking Optimizer States Through Low-Rank Approximation

Stop Tracking Me! Proactive Defense Against Attribute Inference Attack in LLMs

Do MLLMs Really Understand Space? A Mathematical Reasoning Evaluation

To Trust Or Not To Trust Your Vision-Language Model's Prediction

Stay Unique, Stay Efficient: Preserving Model Personality in Multi-Task Merging

Reranker Helps, but Not Enough: Towards Strong Poisoning Attacks Against RAG

Adaptive Jailbreak Defense: A Self-Evolving Framework for Large Language Models

SafeRBench: A Comprehensive Benchmark for Safety Assessment of Large Reasoning Models

LoRA-Pro: Are Low-Rank Adapters Properly Optimized?

Test-Time Immunization: A Universal Defense Framework Against Jailbreaks for (Multimodal) Large Language Models

Personalized Prompt Tuning for Unsupervised Federated Learning