collaborators

8 papers

cs.AR2025

Focus: A Streaming Concentration Architecture for Efficient Vision-Language Models

Chiyue Wei, Cong Guo, Junyao Zhang +8

Vision-Language Models (VLMs) have demonstrated strong performance on tasks such as video captioning and visual question answering. However, their growing scale and video-level inp…

cs.LG2025

FlashSVD: Memory-Efficient Inference with Streaming for Low-Rank Models

Zishan Shao, Yixiao Wang, Qinsi Wang +6

Singular Value Decomposition (SVD) has recently seen a surge of interest as a simple yet powerful tool for large language models (LLMs) compression, with a growing number of works…

cs.CL2025

SafeTy Reasoning Elicitation Alignment for Multi-Turn Dialogues

Martin Kuo, Jianyi Zhang, Aolin Ding +12

Malicious attackers can exploit large language models (LLMs) by engaging them in multi-turn dialogues to achieve harmful objectives, posing significant safety risks to society. To…

cs.LG2025

Keyframe-oriented Vision Token Pruning: Enhancing Efficiency of Large Vision Language Models on Long-Form Video Processing

Yudong Liu, Jingwei Sun, Yueqian Lin +6

Vision language models (VLMs) demonstrate strong capabilities in jointly processing visual and textual data. However, they often incur substantial computational overhead due to red…

cs.CL2025

Proactive Privacy Amnesia for Large Language Models: Safeguarding PII with Negligible Impact on Model Utility

Martin Kuo, Jingyang Zhang, Jianyi Zhang +9

With the rise of large language models (LLMs), increasing research has recognized their risk of leaking personally identifiable information (PII) under malicious attacks. Although…

cs.CL2025

H-CoT: Hijacking the Chain-of-Thought Safety Reasoning Mechanism to Jailbreak Large Reasoning Models, Including OpenAI o1/o3, DeepSeek-R1, and Gemini 2.0 Flash Thinking

Martin Kuo, Jianyi Zhang, Aolin Ding +6

Large Reasoning Models (LRMs) have recently extended their powerful reasoning capabilities to safety checks-using chain-of-thought reasoning to decide whether a request should be a…