Lune

Profiles2021-2026

Xinchao Wang

National University of Singapore

136Papers
104Last author
124Unique co-authors
0First author
0Solo

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

Recent papers136

  1. dParallel: Learnable Parallel Decoding for dLLMsAuthor 5 of 5ICLR 2026 · 64 citations
  2. SparseD: Sparse Attention for Diffusion Language ModelsAuthor 5 of 5ICLR 2026 · 18 citations
  3. Sponge Tool Attack: Stealthy Denial-of-Efficiency against Tool-Augmented Agentic ReasoningAuthor 2 of 2ICML 2026 · 12 citations
  4. PE3R: Perception-Efficient 3D ReconstructionAuthor 3 of 3CVPR 2026 · 9 citations
  5. SpotEdit: Selective Region Editing in Diffusion TransformersAuthor 5 of 5CVPR 2026 · 7 citations
  6. Every Step Counts: Decoding Trajectories as Authorship Fingerprints of dLLMsAuthor 4 of 4ICML 2026 · 4 citations
  7. Hypergraph-State Collaborative Reasoning for Multi-Object TrackingAuthor 5 of 5CVPR 2026 · 4 citations
  8. Invisible Safety Threat: Malicious Finetuning for LLM via SteganographyAuthor 4 of 4ICLR 2026 · 4 citations
  9. Understanding Dataset Distillation via Spectral FilteringAuthor 3 of 3ICLR 2026 · 3 citations
  10. Gated Condition Injection without Multimodal Attention: Towards Controllable Linear-Attention TransformersAuthor 5 of 5CVPR 2026 · 2 citations
  11. MixReasoning: Switching Modes to ThinkAuthor 5 of 5ICML 2026 · 2 citations
  12. Rethinking Token Reduction for Large Vision-Language ModelsAuthor 10 of 11CVPR 2026 · 1 citation
  13. Align Images Before You GenerateAuthor 3 of 3CVPR 2026
  14. Beyond Soft Label: Dataset Distillation via Orthogonal Gradient MatchingAuthor 2 of 2CVPR 2026
  15. CoDA: From Text-to-Image Diffusion Models to Training-Free Dataset DistillationAuthor 3 of 3ICLR 2026
  16. Don't Overthink with Pixels: Efficient Reasoning for SegmentationAuthor 9 of 9ICML 2026
  17. Efficient Multi-modal Dataset Distillation via Analytic Parameter MatchingAuthor 2 of 2ICML 2026
  18. Exploiting Blurry Representations for Event-guided Video Super-ResolutionAuthor 2 of 2AAAI 2026
  19. Flex-Forcing: Towards a Unified Autoregressive and Bidirectional Video Diffusion ModelAuthor 6 of 6ICML 2026
  20. FreLay: Frequency-aware Energy Function for Training-free Layout-to-Image GenerationAuthor 5 of 5AAAI 2026
  21. Gradient Inversion Attacks Beyond SGDAuthor 4 of 4ICML 2026
  22. Language-guided Frequency Modulation for Large Vision-Language ModelsAuthor 6 of 6CVPR 2026
  23. Minute-Long Videos with Dual ParallelismsAuthor 6 of 6AAAI 2026
  24. OminiControl2: Efficient Conditioning for Diffusion TransformersAuthor 5 of 5CVPR Findings 2026
  25. SPEED: Sharpened-Teacher Distillation for Parallel Decoding of Diffusion Language ModelsAuthor 5 of 5ICML 2026
  26. Taming the Phantom: Token-Asymmetric Filtering for Hallucination Mitigation in Large Vision-Language ModelsAuthor 6 of 6AAAI 2026
  27. CoT-Valve: Length-Compressible Chain-of-Thought TuningAuthor 5 of 5ACL 2025 · 162 citations
  28. Thinkless: LLM Learns When to ThinkAuthor 3 of 3NeurIPS 2025 · 128 citations
  29. GFlow: Recovering 4D World from Monocular VideoAuthor 5 of 6AAAI 2025 · 47 citations
  30. OminiControl: Minimal and Universal Control for Diffusion TransformerAuthor 5 of 5ICCV 2025 · 34 citations

Showing the 30 most recent indexed papers.

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