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。
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.
Zhangchen Xu, Fengqing Jiang, Luyao Niu +4 authors
Magpie, a self-synthesis method, generates high-quality instruction data from aligned LLMs, enabling competitive performance in alignment benchmarks compared to models fine-tuned with larger, supervised datasets.
Hanze Dong, Wei Xiong, Bo Pang +7 authors
Online iterative reinforcement learning from human feedback achieves state-of-the-art performance in large language models using open-source datasets and proxy preference models.
Lei Bai, Jiaqi Cao, Chiyu Chen +122 authors
Intern-S2-Preview is a scientific agentic foundation model series that integrates multimodal pre-training, multi-task reinforcement learning, and memory-augmented extensions to support long-horizon scientific reasoning and forecasting.
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.
Zixiang Chen, Yihe Deng, Huizhuo Yuan +2 authors
A novel self-play fine-tuning method (SPIN) enhances Large Language Models by generating its own training data and refining its policy, achieving performance on par with models trained using direct preference optimization and additional human data.
Jiaqi Tang, Jianmin Chen, Wei Wei +7 authors
A novel framework, Robust-R1, enhances multimodal large language models' robustness to visual degradations through explicit modeling, supervised fine-tuning, reward-driven alignment, and dynamic reasoning depth scaling, achieving state-of-the-art performance on real-world degradation benchmarks.
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.
Kang Chen, Zhihao Liu, Tonghe Zhang +10 authors
The framework π<sub>RL</sub> uses reinforcement learning to train flow-based Vision-Language-Action models, addressing challenges with intractable action log-likelihoods and achieving significant performance improvements over supervised fine-tuning.
Kanana LLM Team, Yunju Bak, Hojin Lee +26 authors
Kanana, a series of bilingual language models, achieves superior performance in Korean and competitive performance in English with lower computational costs through efficient pre-training and post-training techniques.
Alexander Amini, Anna Banaszak, Harold Benoit +30 authors
LFM2, a family of compact foundation models, achieves high efficiency and performance on-device through hardware-in-the-loop architecture search and advanced training techniques, supporting various tasks including multimodal applications.
Siheng Li, Cheng Yang, Zesen Cheng +4 authors
A novel method enables LLMs to self-improve in long-context reasoning through sampling, scoring with Minimum Bayes Risk, and supervised fine-tuning, outperforming expert-annotated data approaches.
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.
Yiran Zhao, Chaoqun Liu, Yue Deng +8 authors
Babel is an open multilingual LLM that expands its parameter count through layer extension, covering numerous languages and achieving superior performance in multilingual tasks compared to other open LLMs.
Zhengxi Lu, Yuxiang Chai, Yaxuan Guo +5 authors
Rule-based reinforcement learning enhances multimodal large language models for GUI action prediction, improving data efficiency and performance.
Jiazhen Pan, Che Liu, Junde Wu +6 authors
MedVLM-R1 enhances medical image analysis through explicit natural language reasoning generation, achieving superior accuracy with limited data and parameters via reinforcement learning rather than supervised fine-tuning.
Huajian Xin, Z. Z. Ren, Junxiao Song +14 authors
DeepSeek-Prover-V1.5 improves theorem proving by optimizing training and inference, utilizing reinforcement learning, and proposing RMaxTS for diverse proof paths, achieving state-of-the-art results on miniF2F and ProofNet benchmarks.
Ziyuan Liu, Hengqi Liu, Zichuan Wang +6 authors
Two large-scale search agents are trained via a multi-stage pipeline combining supervised fine-tuning and reinforcement learning against live search, achieving state-of-the-art open-source results on complex web benchmarks through rigorous trajectory filtering and inference-time context management.
Nuo Chen, Zhiyuan Hu, Qingyun Zou +4 authors
JudgeLRM models trained with reinforcement learning outperform existing models, especially in tasks requiring deep reasoning.
Jiahao Zhao, Xiaomin Yu, Zhongxiang Sun +5 authors
ToolArtist is an agentic image generation model that unifies reasoning, tool use, and image synthesis under a single policy via supervised fine-tuning and reinforcement learning.
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