Related papers: Entropy-Gated Selective Policy Optimization:Token-…

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…

Large language models are increasingly trained via reinforcement learning for personalized recommendation tasks, but standard methods like GRPO rely on sparse, sequence-level rewards. These obscure which tokens actually contribute to…

Since DeepSeek-R1 popularized, Group Relative Policy Optimization (GRPO) has become the core part of training Reasoning LLMs. However, we find some deficiency that influences RL stability and inference efficiency, like zero-variance in…

Large Language Models (LLMs) show potential as sequential decision-making agents, but their application is often limited due to a reliance on large, computationally expensive models. This creates a need to improve smaller models, yet…

The remarkable performance of Large Language Models (LLMs) highly relies on crafted prompts. However, manual prompt engineering is a laborious process, creating a core bottleneck for practical application of LLMs. This phenomenon has led to…

Image captioning tasks usually use two-stage training to complete model optimization. The first stage uses cross-entropy as the loss function for optimization, and the second stage uses self-critical sequence training (SCST) for…

Combining Chain-of-Thought (CoT) with Reinforcement Learning (RL) improves text-to-image (T2I) generation, yet the underlying interaction between CoT's exploration and RL's optimization remains unclear. We present a systematic entropy-based…

Agentic Test-Time Scaling (TTS) has delivered state-of-the-art (SOTA) performance on complex software engineering tasks such as code generation and bug fixing. However, its practical adoption remains limited due to significant computational…

Reinforcement learning from verifiable rewards has emerged as a powerful technique for enhancing the complex reasoning abilities of Large Language Models (LLMs). However, these methods are fundamentally constrained by the ''learning cliff''…

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…

Supervised fine-tuning (SFT) followed by reinforcement learning (RL) has become a standard post-training paradigm for large language models. This paradigm provides a cold-start for RL exploration, avoiding the inefficiency of pure RL where…

Reinforcement learning (RL) has emerged as an effective approach for enhancing the reasoning capabilities of large language models (LLMs), especially in scenarios where supervised fine-tuning (SFT) falls short due to limited…

Automatic Prompt Optimization (APO) is a powerful approach for extracting performance from large language models without modifying their weights. Many existing methods rely on trial-and-error, testing different prompts or in-context…

Preference optimization has become a central paradigm for aligning large language models with human feedback. Direct Preference Optimization (DPO) simplifies reinforcement learning from human feedback by directly optimizing pairwise…

Multimodal reward models are crucial for aligning multimodal large language models with human preferences. Recent works have incorporated reasoning capabilities into these models, achieving promising results. However, training these models…

The enhancement of reasoning capabilities in large language models (LLMs) has garnered significant attention, with supervised fine-tuning (SFT) and reinforcement learning emerging as dominant paradigms. While recent studies recognize the…

Hybrid Group Relative Policy Optimization (Hybrid GRPO) is a reinforcement learning framework that extends Proximal Policy Optimization (PPO) and Group Relative Policy Optimization (GRPO) by incorporating empirical multi-sample action…

Large language models (LLMs) exhibit varying levels of confidence across input prompts (questions): some lead to consistent, semantically similar answers, while others yield diverse or contradictory outputs. This variation reflects LLM's…

Reinforcement Learning with Verifiable Rewards (RLVR) offers a promising framework for optimizing large language models in reasoning tasks. However, existing RLVR algorithms focus on different granularities, and each has complementary…

Group-Relative Policy Optimization (GRPO) is a key technique for training large reasoning models, yet it suffers from a critical vulnerability: the \emph{Think-Answer Mismatch}, where noisy reward signals corrupt the learning process. This…