Wang, Lean
- DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning
DeepSeek-R1 shows an LLM can learn strong step-by-step reasoning from pure reinforcement learning, with no human-labeled reasoning examples.
2025/01/22 by DeepSeek-AI, Daya Guo, Dejian Yang +404 · 93 voices · 1658 citations
Computer Science · #Reinforcement Learning in Robotics #Data Stream Mining Techniques #Explainable Artificial Intelligence (XAI)
- Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention
2025/02/16 by Jingyang Yuan, Huazuo Gao, Yuan, Jingyang +28 · 19 voices · 126 citations
Computer Science · #Advanced Neural Network Applications #Stochastic Gradient Optimization Techniques #cs.AI #cs.CL #cs.LG
- DeepSeek-V3 Technical Report
2024/12/27 by DeepSeek-AI, Aixin Liu, Bei Feng +404 · 39 voices · 6 citations
Computer Science · Engineering · #Distributed and Parallel Computing Systems #Robotics and Automated Systems #cs.AI #cs.CL
- DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model
2024/05/07 by Aixin Liu, DeepSeek-AI, Liu, Aixin +310 · 5 voices · 234 citations
Computer Science · #Expert finding and Q&A systems #Topic Modeling #Speech and dialogue systems
- Auxiliary-Loss-Free Load Balancing Strategy for Mixture-of-Experts
2024/08/28 by Lean Wang, Huazuo Gao, Wang, Lean +7 · 51 citations
Computer Science · #Computation and Language (cs.CL) #Distributed Sensor Networks and Detection Algorithms #FOS: Computer and information sciences #Machine Learning (cs.LG) #Mobile Crowdsensing and Crowdsourcing #Privacy-Preserving Technologies in Data
- Label Words are Anchors: An Information Flow Perspective for Understanding In-Context Learning
2023/05/23 by Lean Wang, Lei Li, Wang, Lean +13 · 27 citations
Computer Science · #Topic Modeling #Domain Adaptation and Few-Shot Learning #Multimodal Machine Learning Applications
- DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models
2024/05/31 by Linli Yao, Yao, Linli, Lei Li +11 · 14 citations
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Natural Language Processing Techniques #Topic Modeling
- Towards Codable Watermarking for Injecting Multi-bits Information to LLMs
2023/07/29 by Lean Wang, Wang, Lean, Wenkai Yang +13 · 9 citations
Computer Science · #Advanced Steganography and Watermarking Techniques #Computation and Language (cs.CL) #Digital Media Forensic Detection #FOS: Computer and information sciences #Video Coding and Compression Technologies
- DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
2025/12/02 by DeepSeek-AI, Liu, Aixin, Mei, Aoxue +260 · 36 citations
#Computation and Language (cs.CL) #FOS: Computer and information sciences
- TimeChat-Online: 80% Visual Tokens are Naturally Redundant in Streaming Videos
2025/04/24 by Linli Yao, Yao, Linli, Yicheng Li +25 · 12 citations
Social Sciences · Computer Science · #Multimedia Communication and Technology #Image and Video Quality Assessment #Video Analysis and Summarization
- Temporal Reasoning Transfer from Text to Video
2024/10/08 by Li, Lei, Liu, Yuanxin, Yao, Linli +6 · 6 citations
#Computation and Language (cs.CL) #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences
- Functional Scaling Laws in Kernel Regression: Loss Dynamics and Learning Rate Schedules
2025/09/23 by Binghui Li, Li, Binghui, Fengling Chen +5 · 3 citations
Computer Science · #Topic Modeling #Domain Adaptation and Few-Shot Learning #Speech Recognition and Synthesis