# DeepSeek 官方论文合集（原文 + 中文翻译） 本目录收录 **深度求索（DeepSeek / DeepSeek-AI）自成立以来以论文形式公开发布的全部研究成果**，共 **33 篇**（截至 2026-09-20）。 每篇论文均包含两部分： - **原文 PDF**：从 arXiv 下载的官方预印本（未做改动） - **中文 Markdown**：摘要全文翻译 + 核心贡献 + 方法要点 + 实验结果 + 结论，便于快速掌握论文内容 > 说明：本次中文版本为「摘要 + 核心贡献」精读版。如需某篇论文的全文逐段翻译，可在此目录基础上追加。 ## 收录范围与判定依据 论文清单通过以下渠道交叉核验后确定，凡判定为 DeepSeek 官方成果者全部收录： 1. **DeepSeek 官网「研究索引」**（https://www.deepseek.com/news ）——官方自行列出的论文条目 2. **arXiv 作者检索**：作者字段含 `DeepSeek-AI` 的全部论文 3. **论文署名单位核验**：逐篇抓取 arXiv 摘要页，确认作者单位为 `DeepSeek-AI` / `DeepSeek AI` 4. **社区索引交叉比对**：Hugging Face `DeepSeek Papers` 合集、第三方综述 判定标准：论文作者单位标注为 DeepSeek-AI / DeepSeek AI，或作者主体为 DeepSeek 在职研究员。仅「以 DeepSeek 模型为研究对象」的第三方论文（数量众多）**不予收录**。 > **另有一项非论文形式的研究发布**：[LPLB（Linear-Programming-Based Load Balancer）](https://github.com/deepseek-ai/LPLB)——DeepSeek 于 2025 年 11 月 1 日发布的 MoE 专家并行负载均衡器，以 GitHub 仓库形式发布（早期研究阶段），未上传 arXiv 预印本，故不在本目录论文之列。 ## 目录结构 ``` 大模型/deepseek/ ├── README.md 本索引 ├── 01-基础模型/ … 09-基础设施/ 各篇论文的原文 PDF + 中文 Markdown（同名配对） ├── _text/ 论文正文纯文本（供后续「全文逐段翻译」使用） ├── _digest/ 摘要 + 章节标题 + 正文首尾（精读写作素材） └── _scripts/ 清单核验、PDF 下载、正文提取、索引生成脚本 ``` `_text/` 与 `_digest/` 已按 arXiv 编号逐篇存好论文全文，后续若要把某几篇扩写为**全文逐段翻译**，可直接基于这些文件进行，无需重新抓取。 > **DreamCraft3D**（2310.16818）为 DeepSeek 参与度有限的早期合作论文：作者单位列入 DeepSeek AI，论文注明「实习期间部分完成」，为完整起见一并收录并单独标注。 --- ## 目录 - [基础模型](#01-基础模型)（7 篇） - [代码模型](#02-代码模型)（3 篇） - [数学模型](#03-数学模型)（2 篇） - [推理模型](#04-推理模型)（1 篇） - [多模态](#05-多模态)（8 篇） - [形式化证明](#06-形式化证明)（3 篇） - [架构](#07-架构)（4 篇） - [训练方法](#08-训练方法)（2 篇） - [基础设施](#09-基础设施)（3 篇） --- ## 论文总览 | # | 论文 | arXiv | 发布日期 | 类别 | |---|---|---|---|---| | 1 | [DeepSeek LLM: Scaling Open-Source Language Models with Longtermism](01-基础模型/01-DeepSeek-LLM-2401.02954_abstract.md) | [2401.02954](https://arxiv.org/abs/2401.02954) | 5 Jan 2024 | 基础模型 | | 2 | [DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models](01-基础模型/02-DeepSeekMoE-2401.06066_abstract.md) | [2401.06066](https://arxiv.org/abs/2401.06066) | 11 Jan 2024 | 基础模型 | | 3 | [DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model](01-基础模型/03-DeepSeek-V2-2405.04434_abstract.md) | [2405.04434](https://arxiv.org/abs/2405.04434) | 7 May 2024 | 基础模型 | | 4 | [DeepSeek-V3 Technical Report](01-基础模型/04-DeepSeek-V3-2412.19437_abstract.md) | [2412.19437](https://arxiv.org/abs/2412.19437) | 27 Dec 2024 | 基础模型 | | 5 | [DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models](01-基础模型/05-DeepSeek-V3.2-2512.02556_abstract.md) | [2512.02556](https://arxiv.org/abs/2512.02556) | 2 Dec 2025 | 基础模型 | | 6 | [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](01-基础模型/06-DeepSeek-V4-2606.19348_abstract.md) | [2606.19348](https://arxiv.org/abs/2606.19348) | 26 Apr 2026 | 基础模型 | | 7 | [DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression](01-基础模型/07-DeepSeek-V4.1-Flash-2609.19969_abstract.md) | [2609.19969](https://arxiv.org/abs/2609.19969) | 17 Sep 2026 | 基础模型 | | 8 | [DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence](02-代码模型/01-DeepSeek-Coder-2401.14196_abstract.md) | [2401.14196](https://arxiv.org/abs/2401.14196) | 25 Jan 2024 | 代码模型 | | 9 | [DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence](02-代码模型/02-DeepSeek-Coder-V2-2406.11931_abstract.md) | [2406.11931](https://arxiv.org/abs/2406.11931) | 17 Jun 2024 | 代码模型 | | 10 | [CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction](02-代码模型/03-CodeIO-2502.07316_abstract.md) | [2502.07316](https://arxiv.org/abs/2502.07316) | 11 Feb 2025 | 代码模型 | | 11 | [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](03-数学模型/01-DeepSeekMath-2402.03300_abstract.md) | [2402.03300](https://arxiv.org/abs/2402.03300) | 5 Feb 2024 | 数学模型 | | 12 | [DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning](03-数学模型/02-DeepSeekMath-V2-2511.22570_abstract.md) | [2511.22570](https://arxiv.org/abs/2511.22570) | 27 Nov 2025 | 数学模型 | | 13 | [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](04-推理模型/01-DeepSeek-R1-2501.12948_abstract.md) | [2501.12948](https://arxiv.org/abs/2501.12948) | 22 Jan 2025 | 推理模型 | | 14 | [DeepSeek-VL: Towards Real-World Vision-Language Understanding](05-多模态/01-DeepSeek-VL-2403.05525_abstract.md) | [2403.05525](https://arxiv.org/abs/2403.05525) | 8 Mar 2024 | 多模态 | | 15 | [Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation](05-多模态/02-Janus-2410.13848_abstract.md) | [2410.13848](https://arxiv.org/abs/2410.13848) | 17 Oct 2024 | 多模态 | | 16 | [DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding](05-多模态/03-DeepSeek-VL2-2412.10302_abstract.md) | [2412.10302](https://arxiv.org/abs/2412.10302) | 13 Dec 2024 | 多模态 | | 17 | [Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling](05-多模态/04-Janus-Pro-2501.17811_abstract.md) | [2501.17811](https://arxiv.org/abs/2501.17811) | 29 Jan 2025 | 多模态 | | 18 | [DeepSeek-OCR: Contexts Optical Compression](05-多模态/05-DeepSeek-OCR-2510.18234_abstract.md) | [2510.18234](https://arxiv.org/abs/2510.18234) | 21 Oct 2025 | 多模态 | | 19 | [DeepSeek-OCR 2: Visual Causal Flow](05-多模态/06-DeepSeek-OCR-2-2601.20552_abstract.md) | [2601.20552](https://arxiv.org/abs/2601.20552) | 28 Jan 2026 | 多模态 | | 20 | [JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation](05-多模态/07-JanusFlow-2411.07975_abstract.md) | [2411.07975](https://arxiv.org/abs/2411.07975) | 12 Nov 2024 | 多模态 | | 21 | [DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior](05-多模态/08-DreamCraft3D-2310.16818_abstract.md) | [2310.16818](https://arxiv.org/abs/2310.16818) | 25 Oct 2023 | 多模态 | | 22 | [DeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic Data](06-形式化证明/01-DeepSeek-Prover-2405.14333_abstract.md) | [2405.14333](https://arxiv.org/abs/2405.14333) | 23 May 2024 | 形式化证明 | | 23 | [DeepSeek-Prover-V1.5: Harnessing Proof Assistant Feedback for Reinforcement Learning and Monte-Carlo Tree Search](06-形式化证明/02-DeepSeek-Prover-V1.5-2408.08152_abstract.md) | [2408.08152](https://arxiv.org/abs/2408.08152) | 15 Aug 2024 | 形式化证明 | | 24 | [DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition](06-形式化证明/03-DeepSeek-Prover-V2-2504.21801_abstract.md) | [2504.21801](https://arxiv.org/abs/2504.21801) | 30 Apr 2025 | 形式化证明 | | 25 | [mHC: Manifold-Constrained Hyper-Connections](07-架构/01-mHC-2512.24880_abstract.md) | [2512.24880](https://arxiv.org/abs/2512.24880) | 31 Dec 2025 | 架构 | | 26 | [Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models](07-架构/02-Engram-条件记忆-2601.07372_abstract.md) | [2601.07372](https://arxiv.org/abs/2601.07372) | 12 Jan 2026 | 架构 | | 27 | [Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention](07-架构/03-NSA-原生稀疏注意力-2502.11089_abstract.md) | [2502.11089](https://arxiv.org/abs/2502.11089) | 16 Feb 2025 | 架构 | | 28 | [Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts](07-架构/04-Aux-Loss-Free-Load-Balancing-2408.15664_abstract.md) | [2408.15664](https://arxiv.org/abs/2408.15664) | 28 Aug 2024 | 架构 | | 29 | [Let the Expert Stick to His Last: Expert-Specialized Fine-Tuning for Sparse Architectural Large Language Models](08-训练方法/01-ESFT-专家专精微调-2407.01906_abstract.md) | [2407.01906](https://arxiv.org/abs/2407.01906) | 2 Jul 2024 | 训练方法 | | 30 | [Inference-Time Scaling for Generalist Reward Modeling](08-训练方法/02-DeepSeek-GRM-2504.02495_abstract.md) | [2504.02495](https://arxiv.org/abs/2504.02495) | 3 Apr 2025 | 训练方法 | | 31 | [Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning](09-基础设施/01-Fire-Flyer-AI-HPC-2408.14158_abstract.md) | [2408.14158](https://arxiv.org/abs/2408.14158) | 26 Aug 2024 | 基础设施 | | 32 | [Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures](09-基础设施/02-Insights-into-DeepSeek-V3-2505.09343_abstract.md) | [2505.09343](https://arxiv.org/abs/2505.09343) | 14 May 2025 | 基础设施 | | 33 | [DualPath: Breaking the Storage Bandwidth Bottleneck in Agentic LLM Inference](09-基础设施/03-DualPath-2602.21548_abstract.md) | [2602.21548](https://arxiv.org/abs/2602.21548) | 25 Feb 2026 | 基础设施 | --- ## 基础模型 > 通用大语言模型系列技术报告（LLM / MoE / V2 / V3 / V3.2 / V4 / V4.1-Flash） ### DeepSeek LLM: Scaling Open-Source Language Models with Longtermism - **arXiv**：2401.02954 ｜ **发布**：5 Jan 2024 - **作者**：DeepSeek-AI : Xiao Bi , Deli Chen , Guanting Chen , Shanhuang Chen , Damai Dai , Chengqi Deng , Honghui Ding , Kai Dong , Qiushi Du , Zhe Fu , Huazuo Gao , Kaige Gao , Wenjun Gao , Ruiqi Ge , Kang Guan , Daya Guo , Jianzhong Guo , Guangbo Hao , Zhewen Hao , Ying He , Wenjie Hu , Panpan Huang , Erhan… - **原文 PDF**：[01-DeepSeek-LLM-2401.02954.pdf](01-基础模型/01-DeepSeek-LLM-2401.02954.pdf) - **中文翻译**：[01-DeepSeek-LLM-2401.02954_abstract.md](01-基础模型/01-DeepSeek-LLM-2401.02954_abstract.md) ### DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models - **arXiv**：2401.06066 ｜ **发布**：11 Jan 2024 - **作者**：Damai Dai , Chengqi Deng , Chenggang Zhao , R.X. Xu , Huazuo Gao , Deli Chen , Jiashi Li , Wangding Zeng , Xingkai Yu , Y. Wu , Zhenda Xie , Y.K. Li , Panpan Huang , Fuli Luo , Chong Ruan , Zhifang Sui , Wenfeng Liang - **原文 PDF**：[02-DeepSeekMoE-2401.06066.pdf](01-基础模型/02-DeepSeekMoE-2401.06066.pdf) - **中文翻译**：[02-DeepSeekMoE-2401.06066_abstract.md](01-基础模型/02-DeepSeekMoE-2401.06066_abstract.md) ### DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model - **arXiv**：2405.04434 ｜ **发布**：7 May 2024 - **作者**：DeepSeek-AI , Aixin Liu , Bei Feng , Bin Wang , Bingxuan Wang , Bo Liu , Chenggang Zhao , Chengqi Dengr , Chong Ruan , Damai Dai , Daya Guo , Dejian Yang , Deli Chen , Dongjie Ji , Erhang Li , Fangyun Lin , Fuli Luo , Guangbo Hao , Guanting Chen , Guowei Li , H. Zhang , Hanwei Xu , Hao Yang , Haowei… - **原文 PDF**：[03-DeepSeek-V2-2405.04434.pdf](01-基础模型/03-DeepSeek-V2-2405.04434.pdf) - **中文翻译**：[03-DeepSeek-V2-2405.04434_abstract.md](01-基础模型/03-DeepSeek-V2-2405.04434_abstract.md) ### DeepSeek-V3 Technical Report - **arXiv**：2412.19437 ｜ **发布**：27 Dec 2024 - **作者**：DeepSeek-AI , Aixin Liu , Bei Feng , Bing Xue , Bingxuan Wang , Bochao Wu , Chengda Lu , Chenggang Zhao , Chengqi Deng , Chenyu Zhang , Chong Ruan , Damai Dai , Daya Guo , Dejian Yang , Deli Chen , Dongjie Ji , Erhang Li , Fangyun Lin , Fucong Dai , Fuli Luo , Guangbo Hao , Guanting Chen , Guowei Li… - **原文 PDF**：[04-DeepSeek-V3-2412.19437.pdf](01-基础模型/04-DeepSeek-V3-2412.19437.pdf) - **中文翻译**：[04-DeepSeek-V3-2412.19437_abstract.md](01-基础模型/04-DeepSeek-V3-2412.19437_abstract.md) ### DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models - **arXiv**：2512.02556 ｜ **发布**：2 Dec 2025 - **作者**：DeepSeek-AI , Aixin Liu , Aoxue Mei , Bangcai Lin , Bing Xue , Bingxuan Wang , Bingzheng Xu , Bochao Wu , Bowei Zhang , Chaofan Lin , Chen Dong , Chengda Lu , Chenggang Zhao , Chengqi Deng , Chenhao Xu , Chong Ruan , Damai Dai , Daya Guo , Dejian Yang , Deli Chen , Erhang Li , Fangqi Zhou , Fangyun … - **原文 PDF**：[05-DeepSeek-V3.2-2512.02556.pdf](01-基础模型/05-DeepSeek-V3.2-2512.02556.pdf) - **中文翻译**：[05-DeepSeek-V3.2-2512.02556_abstract.md](01-基础模型/05-DeepSeek-V3.2-2512.02556_abstract.md) ### DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence - **arXiv**：2606.19348 ｜ **发布**：26 Apr 2026 - **作者**：DeepSeek-AI , Anyi Xu , Bangcai Lin , Bing Xue , Bingxuan Wang , Bingzheng Xu , Bochao Wu , Bowei Zhang , Chaofan Lin , Chen Dong , Chenchen Ling , Chengda Lu , Chenggang Zhao , Chengqi Deng , Chengyu Hou , Chenhao Xu , Chenze Shao , Chong Ruan , Conner Sun , Damai Dai , Daya Guo , Dejian Yang , Del… - **原文 PDF**：[06-DeepSeek-V4-2606.19348.pdf](01-基础模型/06-DeepSeek-V4-2606.19348.pdf) - **中文翻译**：[06-DeepSeek-V4-2606.19348_abstract.md](01-基础模型/06-DeepSeek-V4-2606.19348_abstract.md) ### DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression - **arXiv**：2609.19969 ｜ **发布**：17 Sep 2026 - **作者**：DeepSeek-AI : Anyi Xu , B. Li , Bangcai Lin , Bing Xue , BingCheng Xian , Bingzheng Xu , Bochao Wu , Bowei Zhang , Boyi Deng , C.C. Yu , Chao Jin , Chaofan Lin , Chen Dong , Chenbing Wang , Chenfan Feng , Chengda Lu , Chenggang Zhao , Chengqi Deng , Chengyuan Zhang , Chenhao Xu , Chenqi Zhao , Chenz… - **原文 PDF**：[07-DeepSeek-V4.1-Flash-2609.19969.pdf](01-基础模型/07-DeepSeek-V4.1-Flash-2609.19969.pdf) - **中文翻译**：[07-DeepSeek-V4.1-Flash-2609.19969_abstract.md](01-基础模型/07-DeepSeek-V4.1-Flash-2609.19969_abstract.md) ## 代码模型 > 代码智能模型与代码驱动推理（DeepSeek-Coder / Coder-V2 / CodeI/O） ### DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligence - **arXiv**：2401.14196 ｜ **发布**：25 Jan 2024 - **作者**：Daya Guo , Qihao Zhu , Dejian Yang , Zhenda Xie , Kai Dong , Wentao Zhang , Guanting Chen , Xiao Bi , Y. Wu , Y.K. Li , Fuli Luo , Yingfei Xiong , Wenfeng Liang - **原文 PDF**：[01-DeepSeek-Coder-2401.14196.pdf](02-代码模型/01-DeepSeek-Coder-2401.14196.pdf) - **中文翻译**：[01-DeepSeek-Coder-2401.14196_abstract.md](02-代码模型/01-DeepSeek-Coder-2401.14196_abstract.md) ### DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence - **arXiv**：2406.11931 ｜ **发布**：17 Jun 2024 - **作者**：DeepSeek-AI , Qihao Zhu , Daya Guo , Zhihong Shao , Dejian Yang , Peiyi Wang , Runxin Xu , Y. Wu , Yukun Li , Huazuo Gao , Shirong Ma , Wangding Zeng , Xiao Bi , Zihui Gu , Hanwei Xu , Damai Dai , Kai Dong , Liyue Zhang , Yishi Piao , Zhibin Gou , Zhenda Xie , Zhewen Hao , Bingxuan Wang , Junxiao So… - **原文 PDF**：[02-DeepSeek-Coder-V2-2406.11931.pdf](02-代码模型/02-DeepSeek-Coder-V2-2406.11931.pdf) - **中文翻译**：[02-DeepSeek-Coder-V2-2406.11931_abstract.md](02-代码模型/02-DeepSeek-Coder-V2-2406.11931_abstract.md) ### CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction - **arXiv**：2502.07316 ｜ **发布**：11 Feb 2025 - **作者**：Junlong Li , Daya Guo , Dejian Yang , Runxin Xu , Yu Wu , Junxian He - **原文 PDF**：[03-CodeIO-2502.07316.pdf](02-代码模型/03-CodeIO-2502.07316.pdf) - **中文翻译**：[03-CodeIO-2502.07316_abstract.md](02-代码模型/03-CodeIO-2502.07316_abstract.md) ## 数学模型 > 数学推理模型（DeepSeekMath / DeepSeekMath-V2） ### DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models - **arXiv**：2402.03300 ｜ **发布**：5 Feb 2024 - **作者**：Zhihong Shao , Peiyi Wang , Qihao Zhu , Runxin Xu , Junxiao Song , Xiao Bi , Haowei Zhang , Mingchuan Zhang , Y.K. Li , Y. Wu , Daya Guo - **原文 PDF**：[01-DeepSeekMath-2402.03300.pdf](03-数学模型/01-DeepSeekMath-2402.03300.pdf) - **中文翻译**：[01-DeepSeekMath-2402.03300_abstract.md](03-数学模型/01-DeepSeekMath-2402.03300_abstract.md) ### DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning - **arXiv**：2511.22570 ｜ **发布**：27 Nov 2025 - **作者**：Zhihong Shao , Yuxiang Luo , Chengda Lu , Z.Z. Ren , Jiewen Hu , Tian Ye , Zhibin Gou , Shirong Ma , Xiaokang Zhang - **原文 PDF**：[02-DeepSeekMath-V2-2511.22570.pdf](03-数学模型/02-DeepSeekMath-V2-2511.22570.pdf) - **中文翻译**：[02-DeepSeekMath-V2-2511.22570_abstract.md](03-数学模型/02-DeepSeekMath-V2-2511.22570_abstract.md) ## 推理模型 > 强化学习驱动的推理模型（DeepSeek-R1） ### DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning - **arXiv**：2501.12948 ｜ **发布**：22 Jan 2025 - **作者**：DeepSeek-AI , Daya Guo , Dejian Yang , Haowei Zhang , Junxiao Song , Peiyi Wang , Qihao Zhu , Runxin Xu , Ruoyu Zhang , Shirong Ma , Xiao Bi , Xiaokang Zhang , Xingkai Yu , Yu Wu , Z.F. Wu , Zhibin Gou , Zhihong Shao , Zhuoshu Li , Ziyi Gao , Aixin Liu , Bing Xue , Bingxuan Wang , Bochao Wu , Bei Fe… - **原文 PDF**：[01-DeepSeek-R1-2501.12948.pdf](04-推理模型/01-DeepSeek-R1-2501.12948.pdf) - **中文翻译**：[01-DeepSeek-R1-2501.12948_abstract.md](04-推理模型/01-DeepSeek-R1-2501.12948_abstract.md) ## 多模态 > 视觉-语言与统一多模态模型（VL / VL2 / Janus / Janus-Pro / JanusFlow / OCR 系列 / DreamCraft3D） ### DeepSeek-VL: Towards Real-World Vision-Language Understanding - **arXiv**：2403.05525 ｜ **发布**：8 Mar 2024 - **作者**：Haoyu Lu , Wen Liu , Bo Zhang , Bingxuan Wang , Kai Dong , Bo Liu , Jingxiang Sun , Tongzheng Ren , Zhuoshu Li , Hao Yang , Yaofeng Sun , Chengqi Deng , Hanwei Xu , Zhenda Xie , Chong Ruan - **原文 PDF**：[01-DeepSeek-VL-2403.05525.pdf](05-多模态/01-DeepSeek-VL-2403.05525.pdf) - **中文翻译**：[01-DeepSeek-VL-2403.05525_abstract.md](05-多模态/01-DeepSeek-VL-2403.05525_abstract.md) ### Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation - **arXiv**：2410.13848 ｜ **发布**：17 Oct 2024 - **作者**：Chengyue Wu , Xiaokang Chen , Zhiyu Wu , Yiyang Ma , Xingchao Liu , Zizheng Pan , Wen Liu , Zhenda Xie , Xingkai Yu , Chong Ruan , Ping Luo - **原文 PDF**：[02-Janus-2410.13848.pdf](05-多模态/02-Janus-2410.13848.pdf) - **中文翻译**：[02-Janus-2410.13848_abstract.md](05-多模态/02-Janus-2410.13848_abstract.md) ### DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding - **arXiv**：2412.10302 ｜ **发布**：13 Dec 2024 - **作者**：Zhiyu Wu , Xiaokang Chen , Zizheng Pan , Xingchao Liu , Wen Liu , Damai Dai , Huazuo Gao , Yiyang Ma , Chengyue Wu , Bingxuan Wang , Zhenda Xie , Yu Wu , Kai Hu , Jiawei Wang , Yaofeng Sun , Yukun Li , Yishi Piao , Kang Guan , Aixin Liu , Xin Xie , Yuxiang You , Kai Dong , Xingkai Yu , Haowei Zhang … - **原文 PDF**：[03-DeepSeek-VL2-2412.10302.pdf](05-多模态/03-DeepSeek-VL2-2412.10302.pdf) - **中文翻译**：[03-DeepSeek-VL2-2412.10302_abstract.md](05-多模态/03-DeepSeek-VL2-2412.10302_abstract.md) ### Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling - **arXiv**：2501.17811 ｜ **发布**：29 Jan 2025 - **作者**：Xiaokang Chen , Zhiyu Wu , Xingchao Liu , Zizheng Pan , Wen Liu , Zhenda Xie , Xingkai Yu , Chong Ruan - **原文 PDF**：[04-Janus-Pro-2501.17811.pdf](05-多模态/04-Janus-Pro-2501.17811.pdf) - **中文翻译**：[04-Janus-Pro-2501.17811_abstract.md](05-多模态/04-Janus-Pro-2501.17811_abstract.md) ### DeepSeek-OCR: Contexts Optical Compression - **arXiv**：2510.18234 ｜ **发布**：21 Oct 2025 - **作者**：Haoran Wei , Yaofeng Sun , Yukun Li - **原文 PDF**：[05-DeepSeek-OCR-2510.18234.pdf](05-多模态/05-DeepSeek-OCR-2510.18234.pdf) - **中文翻译**：[05-DeepSeek-OCR-2510.18234_abstract.md](05-多模态/05-DeepSeek-OCR-2510.18234_abstract.md) ### DeepSeek-OCR 2: Visual Causal Flow - **arXiv**：2601.20552 ｜ **发布**：28 Jan 2026 - **作者**：Haoran Wei , Yaofeng Sun , Yukun Li - **原文 PDF**：[06-DeepSeek-OCR-2-2601.20552.pdf](05-多模态/06-DeepSeek-OCR-2-2601.20552.pdf) - **中文翻译**：[06-DeepSeek-OCR-2-2601.20552_abstract.md](05-多模态/06-DeepSeek-OCR-2-2601.20552_abstract.md) ### JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation - **arXiv**：2411.07975 ｜ **发布**：12 Nov 2024 - **作者**：Yiyang Ma , Xingchao Liu , Xiaokang Chen , Wen Liu , Chengyue Wu , Zhiyu Wu , Zizheng Pan , Zhenda Xie , Haowei Zhang , Xingkai yu , Liang Zhao , Yisong Wang , Jiaying Liu , Chong Ruan - **原文 PDF**：[07-JanusFlow-2411.07975.pdf](05-多模态/07-JanusFlow-2411.07975.pdf) - **中文翻译**：[07-JanusFlow-2411.07975_abstract.md](05-多模态/07-JanusFlow-2411.07975_abstract.md) ### DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior - **arXiv**：2310.16818 ｜ **发布**：25 Oct 2023 - **作者**：Jingxiang Sun , Bo Zhang , Ruizhi Shao , Lizhen Wang , Wen Liu , Zhenda Xie , Yebin Liu - **原文 PDF**：[08-DreamCraft3D-2310.16818.pdf](05-多模态/08-DreamCraft3D-2310.16818.pdf) - **中文翻译**：[08-DreamCraft3D-2310.16818_abstract.md](05-多模态/08-DreamCraft3D-2310.16818_abstract.md) ## 形式化证明 > 定理证明与形式化数学（DeepSeek-Prover 系列） ### DeepSeek-Prover: Advancing Theorem Proving in LLMs through Large-Scale Synthetic Data - **arXiv**：2405.14333 ｜ **发布**：23 May 2024 - **作者**：Huajian Xin , Daya Guo , Zhihong Shao , Zhizhou Ren , Qihao Zhu , Bo Liu , Chong Ruan , Wenda Li , Xiaodan Liang - **原文 PDF**：[01-DeepSeek-Prover-2405.14333.pdf](06-形式化证明/01-DeepSeek-Prover-2405.14333.pdf) - **中文翻译**：[01-DeepSeek-Prover-2405.14333_abstract.md](06-形式化证明/01-DeepSeek-Prover-2405.14333_abstract.md) ### DeepSeek-Prover-V1.5: Harnessing Proof Assistant Feedback for Reinforcement Learning and Monte-Carlo Tree Search - **arXiv**：2408.08152 ｜ **发布**：15 Aug 2024 - **作者**：Huajian Xin , Z.Z. Ren , Junxiao Song , Zhihong Shao , Wanjia Zhao , Haocheng Wang , Bo Liu , Liyue Zhang , Xuan Lu , Qiushi Du , Wenjun Gao , Qihao Zhu , Dejian Yang , Zhibin Gou , Z.F. Wu , Fuli Luo , Chong Ruan - **原文 PDF**：[02-DeepSeek-Prover-V1.5-2408.08152.pdf](06-形式化证明/02-DeepSeek-Prover-V1.5-2408.08152.pdf) - **中文翻译**：[02-DeepSeek-Prover-V1.5-2408.08152_abstract.md](06-形式化证明/02-DeepSeek-Prover-V1.5-2408.08152_abstract.md) ### DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition - **arXiv**：2504.21801 ｜ **发布**：30 Apr 2025 - **作者**：Z.Z. Ren , Zhihong Shao , Junxiao Song , Huajian Xin , Haocheng Wang , Wanjia Zhao , Liyue Zhang , Zhe Fu , Qihao Zhu , Dejian Yang , Z.F. Wu , Zhibin Gou , Shirong Ma , Hongxuan Tang , Yuxuan Liu , Wenjun Gao , Daya Guo , Chong Ruan - **原文 PDF**：[03-DeepSeek-Prover-V2-2504.21801.pdf](06-形式化证明/03-DeepSeek-Prover-V2-2504.21801.pdf) - **中文翻译**：[03-DeepSeek-Prover-V2-2504.21801_abstract.md](06-形式化证明/03-DeepSeek-Prover-V2-2504.21801_abstract.md) ## 架构 > 基础架构创新（mHC 超连接 / Engram 条件记忆 / NSA 原生稀疏注意力 / 无辅助损失负载均衡） ### mHC: Manifold-Constrained Hyper-Connections - **arXiv**：2512.24880 ｜ **发布**：31 Dec 2025 - **作者**：Zhenda Xie , Yixuan Wei , Huanqi Cao , Chenggang Zhao , Chengqi Deng , Jiashi Li , Damai Dai , Huazuo Gao , Jiang Chang , Kuai Yu , Liang Zhao , Shangyan Zhou , Zhean Xu , Zhengyan Zhang , Wangding Zeng , Shengding Hu , Yuqing Wang , Jingyang Yuan , Lean Wang , Wenfeng Liang - **原文 PDF**：[01-mHC-2512.24880.pdf](07-架构/01-mHC-2512.24880.pdf) - **中文翻译**：[01-mHC-2512.24880_abstract.md](07-架构/01-mHC-2512.24880_abstract.md) ### Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models - **arXiv**：2601.07372 ｜ **发布**：12 Jan 2026 - **作者**：Xin Cheng , Rui Tian , Wangding Zeng , Damai Dai , Qinyu Chen , Bingxuan Wang , Zhenda Xie , Kezhao Huang , Xingkai Yu , Chengqi Deng , Shangyan Zhou , Chenggang Zhao , Zhewen Hao , Yukun Li , Han Zhang , Zhengyan Zhang , Yixu Wei , M.Y Xu , Huishuai Zhang , Dongyan Zhao , Wenfeng Liang - **原文 PDF**：[02-Engram-条件记忆-2601.07372.pdf](07-架构/02-Engram-条件记忆-2601.07372.pdf) - **中文翻译**：[02-Engram-条件记忆-2601.07372_abstract.md](07-架构/02-Engram-条件记忆-2601.07372_abstract.md) ### Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention - **arXiv**：2502.11089 ｜ **发布**：16 Feb 2025 - **作者**：Jingyang Yuan , Huazuo Gao , Damai Dai , Junyu Luo , Liang Zhao , Zhengyan Zhang , Zhenda Xie , Y. X. Wei , Lean Wang , Zhiping Xiao , Yuqing Wang , Chong Ruan , Ming Zhang , Wenfeng Liang , Wangding Zeng - **原文 PDF**：[03-NSA-原生稀疏注意力-2502.11089.pdf](07-架构/03-NSA-原生稀疏注意力-2502.11089.pdf) - **中文翻译**：[03-NSA-原生稀疏注意力-2502.11089_abstract.md](07-架构/03-NSA-原生稀疏注意力-2502.11089_abstract.md) ### Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts - **arXiv**：2408.15664 ｜ **发布**：28 Aug 2024 - **作者**：Lean Wang , Huazuo Gao , Chenggang Zhao , Xu Sun , Damai Dai - **原文 PDF**：[04-Aux-Loss-Free-Load-Balancing-2408.15664.pdf](07-架构/04-Aux-Loss-Free-Load-Balancing-2408.15664.pdf) - **中文翻译**：[04-Aux-Loss-Free-Load-Balancing-2408.15664_abstract.md](07-架构/04-Aux-Loss-Free-Load-Balancing-2408.15664_abstract.md) ## 训练方法 > 训练与对齐方法（ESFT 专家专精微调 / GRM 通用奖励模型） ### Let the Expert Stick to His Last: Expert-Specialized Fine-Tuning for Sparse Architectural Large Language Models - **arXiv**：2407.01906 ｜ **发布**：2 Jul 2024 - **作者**：Zihan Wang , Deli Chen , Damai Dai , Runxin Xu , Zhuoshu Li , Y. Wu - **原文 PDF**：[01-ESFT-专家专精微调-2407.01906.pdf](08-训练方法/01-ESFT-专家专精微调-2407.01906.pdf) - **中文翻译**：[01-ESFT-专家专精微调-2407.01906_abstract.md](08-训练方法/01-ESFT-专家专精微调-2407.01906_abstract.md) ### Inference-Time Scaling for Generalist Reward Modeling - **arXiv**：2504.02495 ｜ **发布**：3 Apr 2025 - **作者**：Zijun Liu , Peiyi Wang , Runxin Xu , Shirong Ma , Chong Ruan , Peng Li , Yang Liu , Yu Wu - **原文 PDF**：[02-DeepSeek-GRM-2504.02495.pdf](08-训练方法/02-DeepSeek-GRM-2504.02495.pdf) - **中文翻译**：[02-DeepSeek-GRM-2504.02495_abstract.md](08-训练方法/02-DeepSeek-GRM-2504.02495_abstract.md) ## 基础设施 > 训练与推理基础设施（Fire-Flyer AI-HPC / 硬件洞察 / DualPath 存储带宽） ### Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning - **arXiv**：2408.14158 ｜ **发布**：26 Aug 2024 - **作者**：Wei An , Xiao Bi , Guanting Chen , Shanhuang Chen , Chengqi Deng , Honghui Ding , Kai Dong , Qiushi Du , Wenjun Gao , Kang Guan , Jianzhong Guo , Yongqiang Guo , Zhe Fu , Ying He , Panpan Huang , Jiashi Li , Wenfeng Liang , Xiaodong Liu , Xin Liu , Yiyuan Liu , Yuxuan Liu , Shanghao Lu , Xuan Lu , X… - **原文 PDF**：[01-Fire-Flyer-AI-HPC-2408.14158.pdf](09-基础设施/01-Fire-Flyer-AI-HPC-2408.14158.pdf) - **中文翻译**：[01-Fire-Flyer-AI-HPC-2408.14158_abstract.md](09-基础设施/01-Fire-Flyer-AI-HPC-2408.14158_abstract.md) ### Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures - **arXiv**：2505.09343 ｜ **发布**：14 May 2025 - **作者**：Chenggang Zhao , Chengqi Deng , Chong Ruan , Damai Dai , Huazuo Gao , Jiashi Li , Liyue Zhang , Panpan Huang , Shangyan Zhou , Shirong Ma , Wenfeng Liang , Ying He , Yuqing Wang , Yuxuan Liu , Y.X. Wei - **原文 PDF**：[02-Insights-into-DeepSeek-V3-2505.09343.pdf](09-基础设施/02-Insights-into-DeepSeek-V3-2505.09343.pdf) - **中文翻译**：[02-Insights-into-DeepSeek-V3-2505.09343_abstract.md](09-基础设施/02-Insights-into-DeepSeek-V3-2505.09343_abstract.md) ### DualPath: Breaking the Storage Bandwidth Bottleneck in Agentic LLM Inference - **arXiv**：2602.21548 ｜ **发布**：25 Feb 2026 - **作者**：Yongtong Wu , Shaoyuan Chen , Yinmin Zhong , Rilin Huang , Yixuan Tan , Wentao Zhang , Liyue Zhang , Shangyan Zhou , Yuxuan Liu , Shunfeng Zhou , Mingxing Zhang , Xin Jin , Panpan Huang - **原文 PDF**：[03-DualPath-2602.21548.pdf](09-基础设施/03-DualPath-2602.21548.pdf) - **中文翻译**：[03-DualPath-2602.21548_abstract.md](09-基础设施/03-DualPath-2602.21548_abstract.md)