Lune

Profiles2019-2026

Kai Han

Huawei Noah's Ark Lab

42Papers
3Last author
63Unique co-authors
4First author
0Solo

First and last author come from byline position, which some venues assign alphabetically.

Recent papers42

  1. PocketLLM: Ultimate Compression of Large Language Models via Meta NetworksAuthor 5 of 5AAAI 2026 · 2 citations
  2. Eve: Efficient Multimodal Vision Language Models with Elastic Visual ExpertsAuthor 5 of 6AAAI 2025 · 16 citations
  3. Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM ReasoningAuthor 2 of 5ICML 2025
  4. Mixture of Lookup ExpertsAuthor 3 of 7ICML 2025
  5. GeminiFusion: Efficient Pixel-wise Multimodal Fusion for Vision TransformerAuthor 3 of 7ICML 2024 · 64 citations
  6. Kangaroo: Lossless Self-Speculative Decoding for Accelerating LLMs via Double Early ExitingAuthor 5 of 6NeurIPS 2024 · 43 citations
  7. Memory-Space Visual Prompting for Efficient Vision-Language Fine-TuningAuthor 5 of 6ICML 2024 · 24 citations
  8. Star-Agents: Automatic Data Optimization with LLM Agents for Instruction TuningAuthor 7 of 8NeurIPS 2024 · 21 citations
  9. Data-efficient Large Vision Models through Sequential AutoregressionAuthor 7 of 8ICML 2024 · 14 citations
  10. An Empirical Study of Scaling Law for Scene Text RecognitionAuthor 5 of 5CVPR 2024 · 11 citations
  11. MemoryFormer : Minimize Transformer Computation by Removing Fully-Connected LayersAuthor 7 of 9NeurIPS 2024 · 8 citations
  12. ParameterNet: Parameters are All You Need for Large-Scale Visual Pretraining of Mobile NetworksAuthor 1 of 4CVPR 2024
  13. Gold-YOLO: Efficient Object Detector via Gather-and-Distribute MechanismAuthor 6 of 7NeurIPS 2023 · 732 citations
  14. One-for-All: Bridge the Gap Between Heterogeneous Architectures in Knowledge DistillationAuthor 3 of 7NeurIPS 2023 · 205 citations
  15. Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis AggregationAuthor 5 of 8ICCV 2023 · 148 citations
  16. Revisit the Power of Vanilla Knowledge Distillation: from Small Scale to Large ScaleAuthor 3 of 6NeurIPS 2023 · 17 citations
  17. Boosting Semantic Segmentation from the Perspective of Explicit Class EmbeddingsAuthor 3 of 5ICCV 2023 · 9 citations
  18. Masked Image Modeling with Local Multi-Scale ReconstructionAuthor 6 of 6CVPR 2023
  19. Network Expansion For Practical Training AccelerationAuthor 3 of 5CVPR 2023
  20. CMT: Convolutional Neural Networks Meet Vision TransformersAuthor 2 of 7CVPR 2022 · 839 citations
  21. Vision GNN: An Image is Worth Graph of NodesAuthor 1 of 5NeurIPS 2022 · 668 citations
  22. GhostNetV2: Enhance Cheap Operation with Long-Range AttentionAuthor 2 of 6NeurIPS 2022 · 634 citations
  23. Patch Slimming for Efficient Vision TransformersAuthor 2 of 7CVPR 2022 · 173 citations
  24. An Image Patch is a Wave: Phase-Aware Vision MLPAuthor 2 of 7CVPR 2022 · 137 citations
  25. Learning Efficient Vision Transformers via Fine-Grained Manifold DistillationAuthor 4 of 8NeurIPS 2022 · 103 citations
  26. Hire-MLP: Vision MLP via Hierarchical RearrangementAuthor 3 of 8CVPR 2022 · 81 citations
  27. Instance-Aware Dynamic Neural Network QuantizationAuthor 3 of 5CVPR 2022 · 38 citations
  28. Redistribution of Weights and Activations for AdderNet QuantizationAuthor 2 of 6NeurIPS 2022 · 14 citations
  29. Accelerating Sparse Convolution with Column Vector-Wise SparsityAuthor 2 of 8NeurIPS 2022 · 12 citations
  30. A Transformer-Based Object Detector with Coarse-Fine Crossing RepresentationsAuthor 3 of 6NeurIPS 2022 · 5 citations

Showing the 30 most recent indexed papers.

Dusk over the sea between two cliffs drawn in fine vertical lines