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。
Adrian Kosowski, Przemysław Uznański, Jan Chorowski +2 authors
BDH, a biologically inspired Large Language Model, combines scale-free network architecture with Hebbian learning to achieve Transformer-like performance while maintaining interpretability.
Jinguo Zhu, Weiyun Wang, Zhe Chen +44 authors
InternVL3 is a multimodal pre-trained language model that jointly learns from both multimodal data and text, improving performance and scalability through advanced techniques and setting a new state-of-the-art in multimodal tasks.
Inclusion AI, Tiwei Bie, Haoxing Chen +15 authors
LLaDA2.0-Uni is a unified discrete diffusion language model that integrates multimodal understanding and generation through a semantic discrete tokenizer, MoE-based backbone, and diffusion decoder, achieving performance comparable to specialized vision-language models while enabling efficient inference and high-fidelity image generation.
Chengshuai Zhao, Zhen Tan, Pingchuan Ma +5 authors
CoT reasoning in LLMs is found to be limited by the distribution discrepancy between training and test data, suggesting it is not a robust form of reasoning.
Yongliang Wu, Yizhou Zhou, Zhou Ziheng +7 authors
Dynamic Fine-Tuning (DFT) improves the generalization of Large Language Models (LLMs) by dynamically rescaling gradients, outperforming standard Supervised Fine-Tuning (SFT) and showing competitive results in offline reinforcement learning.
Tengyue Xu, Zhuoyang Qian, Gaoge Liu +16 authors
Offline knowledge construction through structured methodological graphs enables more reliable and scalable autonomous scientific discovery by reducing reliance on real-time literature processing.
Qizheng Zhang, Changran Hu, Shubhangi Upasani +10 authors
ACE, a framework for adaptive context engineering, enhances LLM applications by preserving detailed knowledge through structured updates, outperforming baselines in agent and domain-specific tasks with reduced adaptation costs.
Ji Zeng, Dayuan Fu, Tiantian Mi +14 authors
Agentic mid-training enables large language models to develop autonomous software engineering capabilities through specialized data synthesis techniques that bridge the gap between static training data and dynamic development environments.
Zhaoyang Yu, Jiayi Zhang, Huixue Su +10 authors
ReCode, a recursive code generation paradigm, unifies high-level planning and low-level action in a single representation, enhancing decision granularity and data efficiency in LLM-based agents.
Khalil Hennara, Sara Chrouf, Mohamed Motaism Hamed +3 authors
A method for integrating a new language into a large language model enhances performance in the target language while preserving existing knowledge.
Wenhan Liu, Xinyu Ma, Weiwei Sun +4 authors
A reasoning-intensive reranker, ReasonRank, achieves state-of-the-art performance in passage ranking tasks by using synthesized training data and a two-stage post-training approach with reinforcement learning.
Guilherme Penedo, Hynek Kydlíček, Loubna Ben allal +5 authors
FineWeb, a 15-trillion token dataset from Common Crawl snapshots, outperforms other open datasets in LLM training, and its educational subset, FineWeb-Edu, significantly improves performance on knowledge and reasoning benchmarks.
Shaoqiu Zhang, Maoquan Wang, Yuling Shi +5 authors
FastContext separates repository exploration from code solving in LLM agents using specialized exploration models that reduce token consumption and improve resolution rates.
Pierre Colombo, Telmo Pessoa Pires, Malik Boudiaf +8 authors
SaulLM-7B, a large language model with 7 billion parameters, excels in legal text comprehension and generation using instructional fine-tuning on a legal corpus.
Shuming Ma, Hongyu Wang, Shaohan Huang +5 authors
BitNet b1.58 2B4T, a 1-bit Large Language Model with 2 billion parameters, matches the performance of full-precision models while improving computational efficiency.
Minjun Kim, Hyeonseok Lim, Hangyeol Yoo +10 authors
A large-scale investigation into constructing a fully open bilingual LLM for Korean using synthetic data demonstrates that such data can sustain pretraining and achieve performance comparable to multilingual baselines.
Xiaomi LLM-Core Team, Bingquan Xia, Bowen Shen +61 authors
MiMo-7B, a large language model optimized for reasoning tasks, enhances pre-training with data mixing and Multi-Token Prediction, and post-training with reinforcement learning on math and programming problems, achieving superior performance over larger models.
Ningning Wang, Xavier Hu, Pai Liu +11 authors
A study on the efficiency-effectiveness trade-off in LLM-driven agent systems identifies optimal agent framework design to reduce costs while maintaining performance.
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.
Zhenran Xu, Xue Yang, Yiyu Wang +7 authors
ComfyUI-Copilot uses a large language model and multi-agent system to enhance the usability and efficiency of the AI-driven art creation platform ComfyUI.
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