Lune

Profiles2019-2026

Yunhe Wang

Huawei Noah's Ark Lab

100Papers
52Last author
90Unique co-authors
0First author
0Solo

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

Recent papers100

  1. GenVidBench: A 6-Million Benchmark for AI-Generated Video DetectionAuthor 8 of 8AAAI 2026 · 13 citations
  2. Multi-Granularity Semantic Revision for Large Language Model DistillationAuthor 10 of 10ACL 2026 · 3 citations
  3. TinySAM: Pushing the Envelope for Efficient Segment Anything ModelAuthor 7 of 8AAAI 2025 · 57 citations
  4. U-REPA: Aligning Diffusion U-Nets to ViTsAuthor 6 of 6NeurIPS 2025 · 30 citations
  5. Eve: Efficient Multimodal Vision Language Models with Elastic Visual ExpertsAuthor 6 of 6AAAI 2025 · 16 citations
  6. CBQ: Cross-Block Quantization for Large Language ModelsAuthor 11 of 11ICLR 2025 · 1 citation
  7. DECO: Unleashing the Potential of ConvNets for Query-based Detection and SegmentationAuthor 4 of 4ICLR 2025
  8. Forest-of-Thought: Scaling Test-Time Compute for Enhancing LLM ReasoningAuthor 5 of 5ICML 2025
  9. LLM Data Selection and Utilization via Dynamic Bi-level OptimizationAuthor 6 of 7ICML 2025
  10. Mixture of Lookup ExpertsAuthor 7 of 7ICML 2025
  11. SlimLLM: Accurate Structured Pruning for Large Language ModelsAuthor 4 of 4ICML 2025
  12. Multiscale Positive-Unlabeled Detection of AI-Generated TextsAuthor 8 of 8ICLR 2024 · 84 citations
  13. U-DiTs: Downsample Tokens in U-Shaped Diffusion TransformersAuthor 6 of 6NeurIPS 2024 · 59 citations
  14. UFineBench: Towards Text-based Person Retrieval with Ultra-fine GranularityAuthor 7 of 8CVPR 2024 · 45 citations
  15. Kangaroo: Lossless Self-Speculative Decoding for Accelerating LLMs via Double Early ExitingAuthor 6 of 6NeurIPS 2024 · 43 citations
  16. SLAB: Efficient Transformers with Simplified Linear Attention and Progressive Re-parameterized Batch NormalizationAuthor 4 of 4ICML 2024 · 40 citations
  17. Image Processing GNN: Breaking Rigidity in Super-ResolutionAuthor 4 of 4CVPR 2024 · 34 citations
  18. Memory-Space Visual Prompting for Efficient Vision-Language Fine-TuningAuthor 6 of 6ICML 2024 · 24 citations
  19. Star-Agents: Automatic Data Optimization with LLM Agents for Instruction TuningAuthor 8 of 8NeurIPS 2024 · 21 citations
  20. Distilling Semantic Priors from SAM to Efficient Image Restoration ModelsAuthor 9 of 9CVPR 2024 · 20 citations
  21. DiJiang: Efficient Large Language Models through Compact KernelizationAuthor 5 of 5ICML 2024 · 12 citations
  22. An Empirical Study of Scaling Law for Scene Text RecognitionAuthor 4 of 5CVPR 2024 · 11 citations
  23. Enhancing Large Language Models through Adaptive TokenizersAuthor 7 of 7NeurIPS 2024 · 11 citations
  24. ExCP: Extreme LLM Checkpoint Compression via Weight-Momentum Joint ShrinkingAuthor 5 of 6ICML 2024 · 11 citations
  25. MemoryFormer : Minimize Transformer Computation by Removing Fully-Connected LayersAuthor 9 of 9NeurIPS 2024 · 8 citations
  26. Rethinking Optimization and Architecture for Tiny Language ModelsAuthor 10 of 10ICML 2024 · 8 citations
  27. ParameterNet: Parameters are All You Need for Large-Scale Visual Pretraining of Mobile NetworksAuthor 2 of 4CVPR 2024
  28. Gold-YOLO: Efficient Object Detector via Gather-and-Distribute MechanismAuthor 7 of 7NeurIPS 2023 · 732 citations
  29. VanillaNet: the Power of Minimalism in Deep LearningAuthor 2 of 4NeurIPS 2023 · 228 citations
  30. One-for-All: Bridge the Gap Between Heterogeneous Architectures in Knowledge DistillationAuthor 6 of 7NeurIPS 2023 · 205 citations

Showing the 30 most recent indexed papers.

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