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。
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.
Bo-Wen Zhang, Junwei He, Wen Wang +5 authors
CoRT improves rubric-based reinforcement learning by using counterfactual replay to assign token-level credit weights within responses, boosting performance without auxiliary scorers.
Dian Zheng, Harry Lee, Manyuan Zhang +4 authors
InterleaveThinker enables interleaved generation capabilities for image generators through a multi-agent pipeline with planner and critic agents, achieving performance comparable to state-of-the-art models while enhancing reasoning benchmarks.
Zhaorun Chen, Zhuokai Zhao, Kai Zhang +15 authors
DreamGym is a unified framework that synthesizes diverse experiences for scalable online RL training, improving agent performance and reducing real-world interactions.
Xixi Wu, Kuan Li, Yida Zhao +11 authors
ReSum, a novel paradigm with periodic context summarization, enhances web agents' performance on knowledge-intensive tasks by overcoming context window limitations, achieving significant improvements over ReAct.
Runpeng Dai, Kaili Huang, Changsung Kang +1 authors
CoGR trains LLMs to generate compact keywords for both queries and items, enabling direct inverted-index retrieval optimized via co-evolving reinforcement learning.
Zanlin Ni, Shenzhi Wang, Yang Yue +8 authors
Arbitrary order generation in diffusion large language models limits reasoning capability by causing premature solution space collapse, making standard policy optimization more effective.
Daniil Plyusov, Alexey Gorbatovski, Boris Shaposhnikov +3 authors
RLVR methods using group sampling suffer from bias toward likely trajectories and missed rare-correct ones; a difficulty-aware advantage scaling technique improves performance on benchmarks without increasing computational cost.
Xiangxin Zhou, Zichen Liu, Haonan Wang +5 authors
A variational reasoning framework treats thinking traces as latent variables, optimizing them through variational inference to improve language model reasoning.
Diankun Wu, Fangfu Liu, Yi-Hsin Hung +1 authors
Spatial-MLLM improves spatial reasoning in multimodal large language models using a dual-encoder architecture with pretrained 2D and 3D structure encoders, achieving state-of-the-art performance on visual spatial tasks.
Zhenyi Liao, Qingsong Xie, Yanhao Zhang +4 authors
The study enhances visual-spatial reasoning in multi-modal large language models through GRPO training using the VSI-100k dataset, demonstrating significant performance improvements over base models.
Yijun Lu, Rui Ye, Jiajun Wang +4 authors
Answer-Backtracked Credit Assignment improves long-horizon search agents by assigning dense step-level rewards via answer-backtracked clue recovery and clue-anchored scoring, yielding strong benchmark results with small models.
Zichen Liu, Changyu Chen, Wenjun Li +5 authors
A study of R1-Zero training identifies pretraining effects on RL performance and proffers Dr. GRPO to enhance token efficiency, achieving superior accuracy on AIME 2024.
Quy-Anh Dang, Chris Ngo
Reinforcement learning enhances reasoning in small language models with limited resources, demonstrating significant performance improvements in mathematical reasoning tasks.
Yuhan Wang, Zhengxi Lu, Yuchen Yan +6 authors
PaperGym converts scientific papers into training environments by separating research questions from evaluation rubrics, enabling reinforcement learning that improves research planning across multiple model sizes.
Chunji Lv, Yangguang Wei, Junlin Liu +6 authors
Persistent Consistency Self-Distillation improves reinforcement learning for language model agents by weighting token-level teacher guidance according to sustained local support, yielding stronger results on interactive tasks.
Olga Tsymboi, Dmitrii Stoianov, Ramil Latypov +11 authors
A smaller self-hosted LLM trained with separate GRPO experts merged via SLERP outperforms a much larger baseline on instruction following, function-calling, and internal tasks while serving half of platform traffic at lower cost.
Shangqing Tu, Daniel Zhang-Li, Yucheng Wang +21 authors
CogEvol is a family of models that generate structured learning artifacts in a single pass using supervised fine-tuning and reinforcement learning with vision-language rewards, achieving high quality with far fewer parameters and lower cost.
Shubham Gandhi, Saurabh Goyal, Kiran Kate +1 authors
DRACO dynamically generates rubrics and redistributes trajectory-level scores into per-step advantages for reinforcement learning without verifiers, improving long-horizon agent performance.
Peixuan Han, Runhui Wang, Ketan Ramaneti +3 authors
Cliff improves reinforcement learning with verifiable rewards by using an off-the-shelf language model to detect the first reasoning error and shaping token-level advantages accordingly.
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