7 papers
From "Aha Moments" to Controllable Thinking: Toward Meta-Cognitive Reasoning in Large Reasoning Models via Decoupled Reasoning and Control
Rui Ha, Rui Pu, Chaozhuo Li +2
Large Reasoning Models (LRMs) can exhibit step-by-step reasoning, reflection, and backtracking, but these behaviors are often unregulated, leading to overthinking. As a result, LRM…
Not All Tokens Are Worth Caching: Learning Semantic-Aware Eviction for LLM Prefix Caches
Shaoke Fang, Ziang Li, Wenfei Wu +3
Prefix caching is a key optimization in Large Language Model (LLM) serving, reusing attention Key-Value (KV) states across requests with shared prompt prefixes to reduce expensive…
How Real is Your Jailbreak? Fine-grained Jailbreak Evaluation with Anchored Reference
Songyang Liu, Chaozhuo Li, Rui Pu +5
Jailbreak attacks present a significant challenge to the safety of Large Language Models (LLMs), yet current automated evaluation methods largely rely on coarse classifications tha…
LANCET: Neural Intervention via Structural Entropy for Mitigating Faithfulness Hallucinations in LLMs
Chenxu Wang, Chaozhuo Li, Pengbo Wang +7
Large Language Models have revolutionized information processing, yet their reliability is severely compromised by faithfulness hallucinations. While current approaches attempt to…
Beyond Surface-Level Detection: Towards Cognitive-Driven Defense Against Jailbreak Attacks via Meta-Operations Reasoning
Rui Pu, Chaozhuo Li, Rui Ha +3
Defending large language models (LLMs) against jailbreak attacks is essential for their safe and reliable deployment. Existing defenses often rely on shallow pattern matching, whic…
Feint and Attack: Attention-Based Strategies for Jailbreaking and Protecting LLMs
Rui Pu, Chaozhuo Li, Rui Ha +5
Jailbreak attack can be used to access the vulnerabilities of Large Language Models (LLMs) by inducing LLMs to generate the harmful content. And the most common method of the attac…