Lune

Profiles2020-2026

Zhi Wang

Tsinghua University

33Papers
22Last author
47Unique co-authors
0First author
0Solo

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

Recent papers33

  1. SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model AccelerationAuthor 9 of 10ICLR 2026 · 60 citations
  2. Sparse ActionGen: Accelerating Diffusion Policy with Real-time PruningAuthor 6 of 6ICML 2026 · 4 citations
  3. VVS: Accelerating Speculative Decoding for Visual Autoregressive Generation via Partial Verification SkippingAuthor 6 of 6CVPR 2026 · 4 citations
  4. Neural Collapse in Test-Time AdaptationAuthor 7 of 7CVPR 2026 · 2 citations
  5. FREE: Uncertainty-Aware Autoregression for Parallel Diffusion TransformersAuthor 5 of 5CVPR Findings 2026
  6. Semantic-Space Exploration and Exploitation in RLVR for LLM ReasoningAuthor 10 of 10ACL Findings 2026
  7. Test-Time Distillation for Continual Model AdaptationAuthor 7 of 7CVPR Findings 2026
  8. Enhancing Implicit Neural Representations via Symmetric Power TransformationAuthor 6 of 6AAAI 2025 · 5 citations
  9. Accelerating Parallel Diffusion Model Serving with Residual CompressionAuthor 8 of 8NeurIPS 2025 · 3 citations
  10. JAQ: Joint Efficient Architecture Design and Low-Bit Quantization with Hardware-Software Co-ExplorationAuthor 11 of 12AAAI 2025 · 3 citations
  11. Music-Aligned Holistic 3D Dance Generation via Hierarchical Motion ModelingAuthor 8 of 8ICCV 2025 · 3 citations
  12. Beyond A Single AI Cluster: A Survey of Decentralized LLM TrainingAuthor 8 of 8EMNLP 2025 · 2 citations
  13. Feature-Based Instance Neighbor Discovery: Advanced Stable Test-Time Adaptation in Dynamic WorldAuthor 7 of 7NeurIPS 2025 · 2 citations
  14. Understanding Bias Terms in Neural RepresentationsAuthor 6 of 6NeurIPS 2025 · 2 citations
  15. γ-FedHT: Stepsize-Aware Hard-Threshold Gradient Compression in Federated LearningAuthor 7 of 7INFOCOM 2025 · 2 citations
  16. DICE: Staleness-Centric Optimizations for Parallel Diffusion MoE InferenceAuthor 7 of 7ICCV 2025 · 1 citation
  17. SizeGS: Size-aware Compression of 3D Gaussian Splatting via Mixed Integer ProgrammingAuthor 10 of 10ACM MM 2025
  18. Procedural Level Generation with Diffusion Models from a Single ExampleAuthor 5 of 5AAAI 2024 · 11 citations
  19. Stable Heterogeneous Treatment Effect Estimation across Out-of-Distribution PopulationsAuthor 5 of 5ICDE 2024 · 2 citations
  20. GIFD: A Generative Gradient Inversion Method with Feature Domain OptimizationAuthor 4 of 5ICCV 2023 · 62 citations
  21. Curriculum Multi-Negative Augmentation for Debiased Video GroundingAuthor 7 of 8AAAI 2023 · 26 citations
  22. JAVP: Joint-Aware Video Processing with Edge-Cloud Collaboration for DNN InferenceAuthor 4 of 4ACM MM 2023 · 23 citations
  23. FSR: A General Frequency-Oriented Framework to Accelerate Image Super-resolution NetworksAuthor 5 of 6AAAI 2023 · 18 citations
  24. DAGC: Data-Aware Adaptive Gradient CompressionAuthor 4 of 4INFOCOM 2023 · 12 citations
  25. SEAM: Searching Transferable Mixed-Precision Quantization Policy through Large Margin RegularizationAuthor 6 of 7ACM MM 2023 · 11 citations
  26. BadTrack: A Poison-Only Backdoor Attack on Visual Object TrackingAuthor 6 of 6NeurIPS 2023 · 7 citations
  27. One-stage Low-resolution Text Recognition with High-resolution Knowledge TransferAuthor 6 of 7ACM MM 2023 · 5 citations
  28. Batch Adaptative Streaming for Video AnalyticsAuthor 6 of 7INFOCOM 2022 · 24 citations
  29. Cost Effective MLaaS Federation: A Combinatorial Reinforcement Learning ApproachAuthor 4 of 4INFOCOM 2022 · 16 citations
  30. bert2BERT: Towards Reusable Pretrained Language ModelsAuthor 7 of 10ACL 2022

Showing the 30 most recent indexed papers.

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