Lune

Profiles2020-2026

Xinghao Chen

Huawei Technologies

34Papers
12Last author
36Unique co-authors
2First author
0Solo

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

Recent papers34

  1. GenVidBench: A 6-Million Benchmark for AI-Generated Video DetectionAuthor 7 of 8AAAI 2026 · 13 citations
  2. Expert Merging: Model Merging with Unsupervised Expert Alignment and Importance-Guided Layer ChunkingAuthor 7 of 7ICLR 2026 · 6 citations
  3. VLM-Pruner: Buffering for Spatial Sparsity in an Efficient VLM Centrifugal Token Pruning ParadigmAuthor 7 of 7CVPR 2026 · 6 citations
  4. PPE: Positional Preservation Embedding for Token Compression in Multimodal Large Language ModelsAuthor 6 of 6ICLR 2026 · 3 citations
  5. SJD-PAC: Accelerating Speculative Jacobi Decoding via Proactive Drafting and Adaptive ContinuationAuthor 5 of 5CVPR 2026 · 1 citation
  6. Entropy-aware Span-Constrained Optimal Transport for Robust Cross-Tokenizer Knowledge DistillationAuthor 7 of 7ICML 2026
  7. VES-RFT: Rewarding Visual Evidence Sensitivity to Mitigate Hallucinations in Large Vision–Language ModelsAuthor 6 of 7CVPR 2026
  8. TinySAM: Pushing the Envelope for Efficient Segment Anything ModelAuthor 8 of 8AAAI 2025 · 57 citations
  9. ViSpec: Accelerating Vision-Language Models with Vision-Aware Speculative DecodingAuthor 5 of 5NeurIPS 2025 · 24 citations
  10. TinyViM: Frequency Decoupling for Tiny Hybrid Vision MambaAuthor 3 of 3ICCV 2025 · 16 citations
  11. Align-KD: Distilling Cross-Modal Alignment Knowledge for Mobile Vision-Language Large Model EnhancementAuthor 4 of 4CVPR 2025
  12. DECO: Unleashing the Potential of ConvNets for Query-based Detection and SegmentationAuthor 1 of 4ICLR 2025
  13. Single Domain Generalization for Few-Shot Counting via Universal Representation MatchingAuthor 5 of 5CVPR 2025
  14. SlimLLM: Accurate Structured Pruning for Large Language ModelsAuthor 2 of 4ICML 2025
  15. TimePro: Efficient Multivariate Long-term Time Series Forecasting with Variable- and Time-Aware Hyper-stateAuthor 4 of 4ICML 2025
  16. GeminiFusion: Efficient Pixel-wise Multimodal Fusion for Vision TransformerAuthor 7 of 7ICML 2024 · 64 citations
  17. SLAB: Efficient Transformers with Simplified Linear Attention and Progressive Re-parameterized Batch NormalizationAuthor 2 of 4ICML 2024 · 40 citations
  18. ExCP: Extreme LLM Checkpoint Compression via Weight-Momentum Joint ShrinkingAuthor 2 of 6ICML 2024 · 11 citations
  19. Less is More: Focus Attention for Efficient DETRAuthor 4 of 5ICCV 2023 · 128 citations
  20. CMT: Convolutional Neural Networks Meet Vision TransformersAuthor 5 of 7CVPR 2022 · 839 citations
  21. Multimodal Token Fusion for Vision TransformersAuthor 2 of 6CVPR 2022 · 214 citations
  22. Hire-MLP: Vision MLP via Hierarchical RearrangementAuthor 4 of 8CVPR 2022 · 81 citations
  23. Random Normalization Aggregation for Adversarial DefenseAuthor 2 of 4NeurIPS 2022 · 23 citations
  24. BiMLP: Compact Binary Architectures for Vision Multi-Layer PerceptronsAuthor 2 of 3NeurIPS 2022 · 10 citations
  25. An Empirical Study of Adder Neural Networks for Object DetectionAuthor 1 of 5NeurIPS 2021 · 22 citations
  26. Towards Stable and Robust AdderNetsAuthor 3 of 4NeurIPS 2021 · 11 citations
  27. Winograd Algorithm for AdderNetAuthor 4 of 6ICML 2021 · 8 citations
  28. Handling Long-tailed Feature Distribution in AdderNetsAuthor 3 of 4NeurIPS 2021 · 3 citations
  29. Distilling Object Detectors via Decoupled FeaturesAuthor 5 of 7CVPR 2021
  30. HourNAS: Extremely Fast Neural Architecture Search Through an Hourglass LensAuthor 3 of 9CVPR 2021

Showing the 30 most recent indexed papers.

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