中文
相关论文

相关论文: Understanding and Preventing Entropy Collapse in R…

200 篇论文

The advancement of Large Reasoning Models (LRMs) has catalyzed a paradigm shift from reactive ``fast thinking'' text generation to systematic, step-by-step ``slow thinking'' reasoning, unlocking state-of-the-art performance in complex…

人工智能 · 计算机科学 2026-05-25 Junyao Yang , Chen Qian , Kun Wang , Linfeng Zhang , Quanshi Zhang , Yong Liu , Dongrui Liu

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a promising approach to improve the reasoning abilities of Large Language Models (LLMs). Among RLVR algorithms, Group Relative Policy Optimization (GRPO) and its variants…

人工智能 · 计算机科学 2026-04-21 Zhaokang Liao , Yingguo Gao , Yi Yang , Yongheng Hu , Jingting Ding

Recent advances in large language models (LLMs) have demonstrated significant potential in hardware design automation, particularly in using natural language to synthesize Register-Transfer Level (RTL) code. Despite this progress, a gap…

机器学习 · 计算机科学 2026-02-26 Jiahe Shi , Zhengqi Gao , Ching-Yun Ko , Duane Boning

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…

Reinforcement learning with verifiable rewards (RLVR) has proven effective in training large reasoning models (LRMs) by leveraging answer-verifiable signals to guide policy optimization, which, however, suffers from high annotation costs.…

Reinforcement Learning with Verifiable Rewards (RLVR) has become a promising paradigm for scaling reasoning capabilities of Large Language Models (LLMs). However, the sparsity of binary verifier rewards often leads to low efficiency and…

机器学习 · 计算机科学 2026-05-22 Shuo Yang , Jinda Lu , Kexin Huang , Chiyu Ma , Shaohang Wei , Yuyang Liu , Guoyin Wang , Jingren Zhou , Li Yuan

Recent advances in large reasoning models (LRMs) demonstrate that sophisticated behaviors such as multi-step reasoning and self-reflection can emerge via reinforcement learning with verifiable rewards~(\textit{RLVR}). However, existing…

机器学习 · 计算机科学 2025-06-24 Jianhao Yan , Yafu Li , Zican Hu , Zhi Wang , Ganqu Cui , Xiaoye Qu , Yu Cheng , Yue Zhang

Vision-Language-Action (VLA) models such as OpenVLA, Octo, and $\pi_0$ have shown strong generalization by leveraging large-scale demonstrations, yet their performance is still fundamentally constrained by the quality and coverage of…

机器学习 · 计算机科学 2025-10-14 Mingyang Lyu , Yinqian Sun , Erliang Lin , Huangrui Li , Ruolin Chen , Feifei Zhao , Yi Zeng

The policy gradient method enjoys the simplicity of the objective where the agent optimizes the cumulative reward directly. Moreover, in the continuous action domain, parameterized distribution of action distribution allows easy control of…

机器学习 · 计算机科学 2022-12-16 Md Masudur Rahman , Yexiang Xue

Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) for Large Language Model (LLM) reasoning have been hindered by a persistent challenge: exploration collapse. The semantic homogeneity of random rollouts often traps…

机器学习 · 计算机科学 2026-01-12 Huilin Deng , Hongchen Luo , Yue Zhu , Long Li , Zhuoyue Chen , Xinghao Zhao , Ming Li , Jihai Zhang , Mengchang Wang , Yang Cao , Yu Kang

Reinforcement Learning with Verifiable Rewards (RLVR) for Multimodal Large Language Models (MLLMs) is highly dependent on high-quality labeled data, which is often scarce and prone to substantial annotation noise in real-world scenarios.…

Long-term training of large language models (LLMs) requires maintaining stable exploration to prevent the model from collapsing into sub-optimal behaviors. Entropy is crucial in this context, as it controls exploration and helps avoid…

机器学习 · 计算机科学 2026-02-03 Kai Yang , Xin Xu , Yangkun Chen , Weijie Liu , Jiafei Lyu , Zichuan Lin , Deheng Ye , Saiyong Yang

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a powerful approach to enhancing the reasoning capabilities of Large Language Models (LLMs), while its mechanisms are not yet well understood. In this work, we undertake a…

RLHF is widely used to align flow-matching text-to-image models with human preferences, but often leads to severe diversity collapse after fine-tuning. In RL, diversity is often assumed to correlate with policy entropy, motivating entropy…

计算机视觉与模式识别 · 计算机科学 2026-05-13 Xiaofeng Tan , Jun Liu , Bin-Bin Gao , Yuanting Fan , Xi Jiang , Chengjie Wang , Hongsong Wang , Feng Zheng

Asynchronous reinforcement learning has become increasingly central to scaling LLM post-training, delivering major throughput gains by decoupling rollout generation from policy updates. However, widely used policy-gradient objectives such…

机器学习 · 计算机科学 2026-03-03 Luke J. Huang , Zhuoyang Zhang , Qinghao Hu , Shang Yang , Song Han

Reinforcement Learning with Verifiable Reward (RLVR) has proven effective for training reasoning-oriented large language models, but existing methods largely assume high-resource settings with abundant training data. In low-resource…

机器学习 · 计算机科学 2026-04-21 Zhanyu Liu , Qingguo Hu , Ante Wang , Chenqing Liu , Zhishang Xiang , Hui Li , Delai Qiu , Jinsong Su

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a pivotal technique for enhancing the reasoning capabilities of Large Language Models (LLMs). However, the de facto practice of mainstream RL algorithms is to treat all…

机器学习 · 计算机科学 2026-05-12 Xincheng Yao , Ruoqi Li , Cheng Chen , Daoxin Zhang , Yi Wu , Yao Hu , Chongyang Zhang

Self-supervised reinforcement learning (RL) presents a promising approach for enhancing the reasoning capabilities of Large Language Models (LLMs) without reliance on expensive human-annotated data. However, we find that existing methods…

人工智能 · 计算机科学 2025-12-16 Bizhe Bai , Hongming Wu , Peng Ye , Tao Chen

As access to high-quality, domain-specific data grows increasingly scarce, multi-epoch training has become a practical strategy for adapting large language models (LLMs). However, autoregressive models often suffer from performance…

计算与语言 · 计算机科学 2025-12-30 Jiapeng Wang , Yiwen Hu , Yanzipeng Gao , Haoyu Wang , Shuo Wang , Hongyu Lu , Jiaxin Mao , Wayne Xin Zhao , Junyi Li , Xiao Zhang

Recent advances in Reinforcement Learning with Verified Reward (RLVR) have driven the emergence of more sophisticated cognitive behaviors in large language models (LLMs), thereby enhancing their reasoning capabilities. However, in prior…

机器学习 · 计算机科学 2025-08-26 Qingbin Li , Rongkun Xue , Jie Wang , Ming Zhou , Zhi Li , Xiaofeng Ji , Yongqi Wang , Miao Liu , Zheming Yang , Minghui Qiu , Jing Yang