Related papers: From Shallow to Deep: Pinning Semantic Intent via …
Following the success of Group Relative Policy Optimization (GRPO) in foundation LLMs, an increasing number of works have sought to adapt GRPO to Visual Large Language Models (VLLMs) for visual perception tasks (e.g., detection and…
While safety alignment for Multimodal Large Language Models (MLLMs) has gained significant attention, current paradigms primarily target malicious intent or situational violations. We propose shifting the safety frontier toward…
Recent advances in reinforcement learning from human feedback (RLHF) and preference optimization have substantially improved the usability, coherence, and safety of large language models. However, recurring behaviors such as performative…
Despite extensive safety-tuning, large language models (LLMs) remain vulnerable to jailbreak attacks via adversarially crafted instructions, reflecting a persistent trade-off between safety and task performance. In this work, we propose…
The soft-thinking paradigm for Large Language Model (LLM) reasoning can outperform the conventional discrete-token Chain-of-Thought (CoT) reasoning in some scenarios, underscoring its research and application value. However, while the…
Large (vision-)language models exhibit remarkable capability but remain highly susceptible to jailbreaking. Existing safety training approaches aim to have the model learn a refusal boundary between safe and unsafe, based on the user's…
Large Language Models (LLMs) are increasingly exposed to adaptive jailbreaking, particularly in high-stakes Chemical, Biological, Radiological, and Nuclear (CBRN) domains. Although streaming probes enable real-time monitoring, they still…
The effective training of Large Language Models (LLMs) for function calling faces a critical challenge: balancing exploration of complex reasoning paths with stable policy optimization. Standard methods like Supervised Fine-Tuning (SFT)…
Post-training LLMs with Reinforcement Learning, specifically Group Relative Policy Optimization (GRPO), has emerged as a paradigm for enhancing mathematical reasoning. However, standard GRPO relies on scalar correctness rewards that are…
Recent research enhances language model reasoning by scaling test-time compute via longer chain-of-thought traces. This often improves accuracy but also introduces redundancy and high computational cost, especially for small language models…
Recent advances in large language models have highlighted the critical need for precise control over model outputs through predefined constraints. While existing methods attempt to achieve this through either direct instruction-response…
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…
Group Relative Policy Optimization(GRPO) has become a cornerstone of modern reinforcement learning alignment, prized for its efficacy in foregoing an explicit value-critic by leveraging reward normalization across sampled trajectory…
Post-training has become central to improving reasoning and alignment in large language models, where critic-free models enable scalable learning from model-generated outputs but lack principled mechanisms to distinguish informative from…
State-of-the-art reasoning models utilize long chain-of-thought (CoT) to solve increasingly complex problems using more test-time computation. In this work, we explore a long CoT setting where the model makes up to K successive attempts at…
Group Relative Policy Optimization (GRPO) has emerged as the de facto Reinforcement Learning (RL) objective driving recent advancements in Multimodal Large Language Models. However, extending this success to open-source multimodal…
Recent Progress in post-training flow matching for text-to-image (T2I) generation with Group Relative Policy Optimization (GRPO) has demonstrated strong potential. However, it is hindered by a critical limitation: inaccurate advantage…
Reinforcement learning (RL) has emerged as a powerful framework for improving the reasoning capabilities of large language models (LLMs). However, most existing RL approaches rely on sparse outcome rewards, which fail to credit correct…
Group Relative Policy Optimization (GRPO) has emerged as an effective and lightweight framework for post-training visual generative models. However, its performance is fundamentally limited by the ambiguity of textual visual correspondence:…
Reinforcement learning with verifiable rewards helps reasoning, but GRPO-style methods stall on hard prompts where all sampled rollouts fail. SORT adds a repair update for those failures without changing rollout generation: it derives a…