5 papers
LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion
Guanghao Zhou, Panjia Qiu, Cen Chen +4
The safety mechanisms of large language models (LLMs) exhibit notable fragility, as even fine-tuning on datasets without harmful content may still undermine their safety capabiliti…
Shadows in the Code: Exploring the Risks and Defenses of LLM-based Multi-Agent Software Development Systems
Xiaoqing Wang, Keman Huang, Bin Liang +2
The rapid advancement of Large Language Model (LLM)-driven multi-agent systems has significantly streamlined software developing tasks, enabling users with little technical experti…
Mini-Omni-Reasoner: Token-Level Thinking-in-Speaking in Large Speech Models
Zhifei Xie, Ziyang Ma, Zihang Liu +7
Reasoning is essential for effective communication and decision-making. While recent advances in LLMs and MLLMs have shown that incorporating explicit reasoning significantly impro…
Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey
Jason Zhu, Hongyu Li
Large reasoning models (LRMs) like OpenAI o1 and DeepSeek R1 have demonstrated impressive performance on complex reasoning tasks like mathematics and programming with long Chain-of…
Revisiting Catastrophic Forgetting in Large Language Model Tuning
Hongyu Li, Liang Ding, Meng Fang +1
Catastrophic Forgetting (CF) means models forgetting previously acquired knowledge when learning new data. It compromises the effectiveness of large language models (LLMs) during f…