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。
Yafu Li, Runzhe Zhan, Haoran Zhang +25 authors
A systematic approach transforms post-trained reasoning models into rigorous olympiad-level solvers through reverse-perplexity curriculum, two-stage reinforcement learning, and test-time scaling, achieving gold-medal performance on mathematical and physics competitions.
Mingjie Liu, Shizhe Diao, Ximing Lu +5 authors
Prolonged reinforcement learning training (ProRL) uncovers novel reasoning strategies in language models, outperforming base models and suggesting meaningful expansion of reasoning capabilities.
Aviral Kumar, Vincent Zhuang, Rishabh Agarwal +15 authors
The multi-turn online reinforcement learning approach SCoRe improves LLM self-correction using self-generated data and regularization, achieving state-of-the-art performance.
Dong Chen, Fangyun Wei, Ziyu Wan +18 authors
Lens is a compact 3.8B-parameter text-to-image model achieving superior performance with reduced training compute through dense caption datasets, multi-resolution batching, efficient architecture, and optimization techniques.
Junying Chen, Zhenyang Cai, Ke Ji +5 authors
The proposed approach enhances medical reasoning in large language models using verifiable problems, guided search, and reinforcement learning to achieve superior performance in medical problem-solving.
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.
Yuchen Fan, Kaiyan Zhang, Heng Zhou +15 authors
LLMs can serve as efficient simulators for RL tasks by leveraging internal knowledge, reducing reliance on external search engines and improving sim-to-real transfer.
Yingzhe Peng, Gongrui Zhang, Miaosen Zhang +7 authors
A two-stage framework enhances reasoning in large multimodal models by first strengthening text-only reasoning and then generalizing it to multimodal tasks.
Chen Hu, Haikuo Du, Heng Wang +64 authors
Step-DeepResearch, an end-to-end agent enhanced with a data synthesis strategy and progressive training, achieves expert-level capabilities in deep research scenarios, outperforming established models.
Zhenghai Xue, Longtao Zheng, Qian Liu +4 authors
SimpleTIR stabilizes multi-turn Tool-Integrated Reasoning training by filtering out void turns, achieving state-of-the-art performance on math reasoning benchmarks.
Yi-Fan Zhang, Xingyu Lu, Shukang Yin +17 authors
Thyme, a novel paradigm, enables MLLMs to autonomously perform image manipulations and computations, enhancing performance in perception and reasoning tasks through a two-stage training strategy and GRPO-ATS algorithm.
Haozhan Li, Yuxin Zuo, Jiale Yu +18 authors
SimpleVLA-RL, an RL framework for VLA models, enhances long-horizon action planning, achieves state-of-the-art performance, and discovers novel patterns during training.
Kaituo Feng, Kaixiong Gong, Bohao Li +5 authors
Video-R1, leveraging rule-based reinforcement learning and temporal information, enhances video reasoning in multimodal large language models using a combination of video and image data.
Egor Cherepanov, Nikita Kachaev, Artem Zholus +2 authors
The paper provides a framework for understanding and evaluating different types of memory in reinforcement learning agents, using cognitive science-inspired definitions and a standardized experimental methodology.
Mistral-AI, Abhinav Rastogi, Albert Q. Jiang +97 authors
Magistral, a scalable reinforcement learning pipeline, demonstrates that RL can enhance multimodal understanding and instruction following in large language models without requiring existing RL traces.
Jiazhan Feng, Shijue Huang, Xingwei Qu +6 authors
ReTool, a tool-integrated learning framework, enhances reasoning models with real-time code execution and reinforcement learning, significantly improving performance in structured problem-solving tasks like mathematical reasoning.
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.
Sheng-Chieh Lin, Luyu Gao, Barlas Oguz +4 authors
Proposes factuality-aware alignment methods for pre-trained language models to enhance factual accuracy during supervised fine-tuning and reinforcement learning, reducing hallucination.
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