相关论文: SafeGRPO: Self-Rewarded Multimodal Safety Alignmen…

Aligning large language models (LLMs) with human values and safety constraints is challenging, especially when objectives like helpfulness, truthfulness, and avoidance of harm conflict. Reinforcement Learning from Human Feedback (RLHF) has…

Multimodal Large Reasoning Models (MLRMs) demonstrate impressive cross-modal reasoning but often amplify safety risks under adversarial or unsafe prompts, a phenomenon we call the \textit{Reasoning Tax}. Existing defenses mainly act at the…

Vision-Language Models (VLMs) have achieved remarkable progress in multimodal reasoning tasks through enhanced chain-of-thought capabilities. However, this advancement also introduces novel safety risks, as these models become increasingly…

Multimodal Large Language Models (MLLMs) are susceptible to the implicit reasoning risk, wherein innocuous unimodal inputs synergistically assemble into risky multimodal data that produce harmful outputs. We attribute this vulnerability to…

Multimodal large language models (MLLMs) have shown promising capabilities in reasoning tasks, yet still struggle with complex problems requiring explicit self-reflection and self-correction, especially compared to their unimodal text-based…

Reinforcement learning with verifiable reward has recently emerged as a central paradigm for post-training large language models (LLMs); however, prevailing mean-based methods, such as Group Relative Policy Optimization (GRPO), suffer from…

Current safety alignment techniques for large language models (LLMs) face two key challenges: (1) under-generalization, which leaves models vulnerable to novel jailbreak attacks, and (2) over-alignment, which leads to the excessive refusal…

Recent studies generally enhance MLLMs' reasoning capabilities via supervised fine-tuning on high-quality chain-of-thought reasoning data, which often leads models to merely imitate successful reasoning paths without understanding what the…

Reinforcement learning from verifiable rewards (RLVR), especially with Group Relative Policy Optimization (GRPO), has shown strong potential for improving the reasoning capabilities of large vision-language models (LVLMs). However, in…

Large Language Models (LLMs) have shown promise in solving complex mathematical problems, yet they still fall short of producing accurate and consistent solutions. Reinforcement Learning (RL) is a framework for aligning these models with…

While Multimodal Large Language Models (MLLMs) have made remarkable progress in vision-language reasoning, they are also more susceptible to producing harmful content compared to models that focus solely on text. Existing defensive…

Improving and understanding the training dynamics and reasoning of Large Language Models (LLMs) has become essential for their deployment in AI-based security tools, such as software vulnerability detection. In this work, we present an…

Balancing helpfulness and safety (harmlessness) is a critical challenge in aligning large language models (LLMs). Current approaches often decouple these two objectives, training separate preference models for helpfulness and safety, while…

Large language models (LLMs) have become increasingly central to AI applications worldwide, necessitating robust multilingual safety alignment to ensure secure deployment across diverse linguistic contexts. Existing preference learning…

Multimodal LLMs often produce fluent yet unreliable reasoning, exhibiting weak step-to-step coherence and insufficient visual grounding, largely because existing alignment approaches supervise only the final answer while ignoring the…

Group relative policy optimization (GRPO) has become a standard post-training paradigm for improving reasoning and preference alignment in large language models (LLMs), and has recently shown strong effectiveness in LLM-based recommender…

Traditional autonomous driving systems often struggle to connect high-level reasoning with low-level control, leading to suboptimal and sometimes unsafe behaviors. Recent advances in multimodal large language models (MLLMs), which process…

Multimodal Large Language Models (MLLMs) pose critical safety challenges, as they are susceptible not only to adversarial attacks such as jailbreaking but also to inadvertently generating harmful content for benign users. While internal…

The Group Relative Policy Optimization (GRPO) algorithm has demonstrated considerable success in enhancing the reasoning capabilities of large language models (LLMs), as evidenced by DeepSeek-R1. However, the absence of intermediate…

The rapid advancement of multi-modal large reasoning models (MLRMs) -- enhanced versions of multimodal language models (MLLMs) equipped with reasoning capabilities -- has revolutionized diverse applications. However, their safety…