14 papers
Beyond Reasoning Gains: Mitigating General-Capability Forgetting in Large Reasoning Models
Hoang Phan, Xianjun Yang, Yuanshun Yao +6
Reinforcement learning with verifiable rewards (RLVR) has delivered impressive gains in mathematical and multimodal reasoning and has become a standard post-training paradigm for c…
Verifying Chain-of-Thought Reasoning via Its Computational Graph
Zheng Zhao, Yeskendir Koishekenov, Xianjun Yang +2
Current Chain-of-Thought (CoT) verification methods predict reasoning correctness based on outputs (black-box) or activations (gray-box), but offer limited insight into why a compu…
Dr. Zero: Self-Evolving Search Agents without Training Data
Zhenrui Yue, Kartikeya Upasani, Xianjun Yang +5
As high-quality data becomes increasingly difficult to obtain, self-evolution without curated training data has emerged as a promising paradigm. This approach allows large language…
Your thoughts tell who you are: Characterize the reasoning patterns of LRMs
Yida Chen, Yuning Mao, Xianjun Yang +7
Current comparisons of large reasoning models (LRMs) focus on macro-level statistics such as task accuracy or reasoning length. Whether different LRMs reason differently remains an…
Many-Turn Jailbreaking
Xianjun Yang, Liqiang Xiao, Shiyang Li +5
Current jailbreaking work on large language models (LLMs) aims to elicit unsafe outputs from given prompts. However, it only focuses on single-turn jailbreaking targeting one speci…
Weak-to-Strong Jailbreaking on Large Language Models
Xuandong Zhao, Xianjun Yang, Tianyu Pang +4
Large language models (LLMs) are vulnerable to jailbreak attacks - resulting in harmful, unethical, or biased text generations. However, existing jailbreaking methods are computati…