DeepSeek-OCR 2: Visual Causal Flow
Paper·Haoran Wei, Yaofeng Sun, Yukun Li·1/28/2026
Research LabManaged by Attendemia
DeepSeek🇨🇳
A Chinese AI research lab known for developing high-performance, cost-effective, open-weights LLMs like DeepSeek-R1 and V3, which rival top U.S. models in reasoning, coding, and math.
Sign in to followFollowers: 550
Timeline
Page 1Source ↗
mHC: Manifold-Constrained Hyper-Connections
Paper·Zhenda Xie, Yixuan Wei, Huanqi Cao, Chenggang Zhao, Cheng…·1/5/2026
DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
Paper·DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao…·1/4/2026
Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI Architectures
Paper·Chenggang Zhao, Chengqi Deng, Chong Ruan, Damai Dai, Huaz…·12/23/2025
DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
Paper·DeepSeek-AI, Aixin Liu, Aoxue Mei, Bangcai Lin, Bing Xue,…·12/2/2025
DeepSeekMath-V2: Towards Self-Verifiable Mathematical Reasoning
Paper·Zhihong Shao, Yuxiang Luo, Chengda Lu, Z. Z. Ren, Jiewen …·11/27/2025
DeepSeek-OCR: Contexts Optical Compression
Paper·Haoran Wei, Yaofeng Sun, Yukun Li·10/21/2025
Inference-Time Scaling for Generalist Reward Modeling
Paper·Zijun Liu, Peiyi Wang, Runxin Xu, Shirong Ma, Chong Ruan,…·9/25/2025
DeepSeek-Prover-V2: Advancing Formal Mathematical Reasoning via Reinforcement Learning for Subgoal Decomposition
Paper·Z. Z. Ren, Zhihong Shao, Junxiao Song, Huajian Xin, Haoch…·7/18/2025
CodeI/O: Condensing Reasoning Patterns via Code Input-Output Prediction
Paper·Junlong Li, Daya Guo, Dejian Yang, Runxin Xu, Yu Wu, Junx…·5/21/2025
JanusFlow: Harmonizing Autoregression and Rectified Flow for Unified Multimodal Understanding and Generation
Paper·Yiyang Ma, Xingchao Liu, Xiaokang Chen, Wen Liu, Chengyue…·3/24/2025
Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention
Paper·Jingyang Yuan, Huazuo Gao, Damai Dai, Junyu Luo, Liang Zh…·2/27/2025
DeepSeek-V3 Technical Report
Paper·DeepSeek-AI, Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang…·2/18/2025
Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scaling
Paper·Xiaokang Chen, Zhiyu Wu, Xingchao Liu, Zizheng Pan, Wen L…·1/29/2025
DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understanding
Paper·Zhiyu Wu, Xiaokang Chen, Zizheng Pan, Xingchao Liu, Wen L…·12/13/2024
Janus: Decoupling Visual Encoding for Unified Multimodal Understanding and Generation
Paper·Chengyue Wu, Xiaokang Chen, Zhiyu Wu, Yiyang Ma, Xingchao…·10/17/2024
Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning
Paper·Wei An, Xiao Bi, Guanting Chen, Shanhuang Chen, Chengqi D…·8/31/2024
Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts
Paper·Lean Wang, Huazuo Gao, Chenggang Zhao, Xu Sun, Damai Dai·8/28/2024
DeepSeek-Prover-V1.5: Harnessing Proof Assistant Feedback for Reinforcement Learning and Monte-Carlo Tree Search
Paper·Huajian Xin, Z. Z. Ren, Junxiao Song, Zhihong Shao, Wanji…·8/15/2024
Let the Expert Stick to His Last: Expert-Specialized Fine-Tuning for Sparse Architectural Large Language Models
Paper·Zihan Wang, Deli Chen, Damai Dai, Runxin Xu, Zhuoshu Li, …·7/5/2024
Page 1Next