Trends · 研究趋势
数据来自 Hugging Face 论文的 AI 提取关键词,按月统计研究方向的增长与热度。
Chujie Zheng, Shixuan Liu, Mingze Li +9 authors
Group Sequence Policy Optimization (GSPO) is a reinforcement learning algorithm that improves training efficiency and performance of large language models by using sequence-level importance ratios and operations.
Mind Lab, Song Cao, Vic Cao +59 authors
MinT is a managed infrastructure system that enables efficient low-rank adaptation training and serving by keeping base models resident and moving lightweight adapter revisions, scaling across multiple dimensions including large model architectures, reduced storage requirements, and distributed policy management.
Yijiang Li, Bingyang Wang, Yijun Liang +3 authors
Unsupervised on-policy self-distillation improves large language models by using internal consistency and majority-vote pseudo-solutions to correct confident errors without external supervision.
Changhai Zhou, Kieran Liu, Yuhua Zhou +17 authors
LongStraw is an execution framework that enables million-token reinforcement learning post-training with fixed GPU resources by reducing live training graphs through replay and selective state retention.
Fengkai Yang, Zherui Chen, Xiaohan Wang +10 authors
Group-based reinforcement learning from verifier rewards suffers from biased advantage estimation that underestimates hard prompts and overestimates easy prompts, which is addressed through a history-aware adaptive difficulty weighting method that improves performance on mathematical reasoning benchmarks.
Guochao Jiang, Wenfeng Feng, Guofeng Quan +4 authors
A curriculum reinforcement learning framework dynamically adjusts training sample difficulty based on reward variance, improving LLM performance on mathematical reasoning tasks.
Xuekai Zhu, Daixuan Cheng, Dinghuai Zhang +20 authors
FlowRL enhances LLM reinforcement learning by matching the full reward distribution through flow balancing, improving diversity and performance over reward-maximizing methods.
Junkang Wu, Kexin Huang, Jiancan Wu +3 authors
Quantile Advantage Estimation stabilizes reinforcement learning with verifiable rewards by addressing entropy issues and improving performance on large language models.
Shiyin Lu, Yang Li, Yu Xia +39 authors
Ovis2.5, a native-resolution vision transformer with multimodal reasoning, achieves state-of-the-art performance on various benchmarks through advanced training techniques and efficient scaling methods.
Zhengxi Lu, Zhiyuan Yao, Zhuowen Han +8 authors
SDAR enhances reinforcement learning for multi-turn agent training by integrating self-distillation through a sigmoid gate that selectively strengthens positive token-level guidance while mitigating negative teacher rejections.
Yunheng Li, Guohong Mu, Hao Li +4 authors
OraRL improves reinforcement learning post-training for video multimodal language models by integrating oracle rollouts with decoupled advantage estimation and sign-balanced pruning, achieving higher sample efficiency and scalability without chain-of-thought generation.
Shuang Chen, Kaituo Feng, Hangting Chen +7 authors
OpenSearch-VL presents an open-source framework for training advanced multimodal search agents using reinforcement learning, featuring specialized data curation, diverse tool environments, and a novel training algorithm that improves performance across multiple benchmarks.
Zi-Han Wang, Zhengxi Lu, Zhiyuan Yao +10 authors
AgentOPSD uses recursive Bayesian belief updates over turn-level log-probability gaps to assign credit in long-horizon agentic reinforcement learning without extra critics or rollouts.
Zhen Fang, Wenxuan Huang, Yu Zeng +8 authors
Flow-OPD addresses limitations in Flow Matching text-to-image models through a two-stage alignment approach combining on-policy distillation and manifold anchor regularization, achieving significant improvements in generation quality and alignment metrics.
Yiping Wang, Qing Yang, Zhiyuan Zeng +11 authors
Reinforcement learning with verifiable reward using one training example significantly enhances math reasoning capabilities of large language models.
Yibin Wang, Zhimin Li, Yuhang Zang +4 authors
A unified multimodal reward model that incorporates long chains of thought reasoning improves reliability and accuracy in vision reward tasks through exploration-driven reinforcement learning.
Zheyuan Yang, Lyuhao Chen, Arman Cohan +1 authors
Two post-training strategies, distillation and RLVR, enable inference-time scaling in table reasoning tasks, resulting in a model (Table-R1-Zero) that matches GPT-4.1's performance using fewer parameters and shows strong generalization.
Minki Kang, Shizhe Diao, Ryo Hachiuma +4 authors
Agents using vision-language models with extended reasoning face challenges in tool utilization, which are addressed through AXPO, a method that improves performance by optimizing thinking prefixes and tool call resampling.
Zichen Liu, Anya Sims, Keyu Duan +16 authors
GEM, an open-source environment simulator, facilitates experience-based learning for large language models by providing a standardized framework and diverse environments for training and benchmarking reinforcement learning algorithms.
Yibin Wang, Zhimin Li, Yuhang Zang +6 authors
Pref-GRPO, a pairwise preference reward-based GRPO method, enhances text-to-image generation by mitigating reward hacking and improving stability, while UniGenBench provides a comprehensive benchmark for evaluating T2I models.
北京市昌平区探索星信息技术及软件开发工作室
京ICP备2026059466号