Lune

Profiles2020-2026

Yu Wang

Tsinghua University

33Papers
28Last author
41Unique co-authors
0First author
0Solo

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

Recent papers33

  1. Cache-to-Cache: Direct Semantic Communication Between Large Language ModelsAuthor 7 of 7ICLR 2026 · 59 citations
  2. Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language ModelsAuthor 6 of 6ICML 2026 · 20 citations
  3. PM-KVQ: Progressive Mixed-precision KV Cache Quantization for Long-CoT LLMsAuthor 10 of 10ICLR 2026 · 18 citations
  4. MARSHAL: Incentivizing Multi-Agent Reasoning via Self-Play with Strategic LLMsAuthor 13 of 13ICLR 2026 · 15 citations
  5. VS-Bench: Evaluating VLMs for Strategic Abilities in Multi-Agent EnvironmentsAuthor 10 of 10CVPR 2026 · 3 citations
  6. Efficient and Adaptable Overlapping for Computation and Communication via Signaling and ReorderingAuthor 12 of 12EuroSys 2026 · 2 citations
  7. R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token RoutingAuthor 9 of 9NeurIPS 2025 · 32 citations
  8. UniNDP: A Unified Compilation and Simulation Tool for Near DRAM Processing ArchitecturesAuthor 9 of 10HPCA 2025 · 9 citations
  9. PARO: Hardware-Software Co-design with Pattern-aware Reorder-based Attention Quantization in Video Generation ModelsAuthor 9 of 9DAC 2025 · 2 citations
  10. MBQ: Modality-Balanced Quantization for Large Vision-Language ModelsAuthor 13 of 13CVPR 2025
  11. Skeleton-of-Thought: Prompting LLMs for Efficient Parallel GenerationAuthor 6 of 6ICLR 2024 · 105 citations
  12. Evaluating Quantized Large Language ModelsAuthor 9 of 9ICML 2024 · 88 citations
  13. FEASTA: A Flexible and Efficient Accelerator for Sparse Tensor Algebra in Machine LearningAuthor 13 of 13ASPLOS 2024 · 16 citations
  14. Rad-NeRF: Ray-decoupled Training of Neural Radiance FieldAuthor 8 of 8NeurIPS 2024 · 10 citations
  15. TorchSparse++: Efficient Training and Inference Framework for Sparse Convolution on GPUsAuthor 8 of 9MICRO 2023 · 32 citations
  16. An Efficient Accelerator for Point-based and Voxel-based Point Cloud Neural NetworksAuthor 7 of 7DAC 2023 · 25 citations
  17. CogDL: A Comprehensive Library for Graph Deep LearningAuthor 19 of 19WWW 2023 · 25 citations
  18. DF-GAS: a Distributed FPGA-as-a-Service Architecture towards Billion-Scale Graph-based Approximate Nearest Neighbor SearchAuthor 11 of 11MICRO 2023 · 24 citations
  19. Processing-In-Hierarchical-Memory Architecture for Billion-Scale Approximate Nearest Neighbor SearchAuthor 7 of 7DAC 2023 · 15 citations
  20. PIM-HLS: An Automatic Hardware Generation Tool for Heterogeneous Processing-In-Memory-based Neural Network AcceleratorsAuthor 8 of 8DAC 2023 · 9 citations
  21. Memory-Oriented Structural Pruning for Efficient Image RestorationAuthor 9 of 9AAAI 2023 · 6 citations
  22. Dynamic Ensemble of Low-Fidelity Experts: Mitigating NAS "Cold-Start"Author 10 of 10AAAI 2023 · 5 citations
  23. Memory-Efficient and Real-Time SPAD-based dToF Depth Sensor with Spatial and Statistical CorrelationAuthor 10 of 10DAC 2023 · 1 citation
  24. FedCor: Correlation-Based Active Client Selection Strategy for Heterogeneous Federated LearningAuthor 5 of 7CVPR 2022 · 120 citations
  25. DIMMining: pruning-efficient and parallel graph mining on near-memory-computingAuthor 9 of 9ISCA 2022 · 56 citations
  26. Heuristic adaptability to input dynamics for SpMM on CPUsAuthor 9 of 9DAC 2022 · 27 citations
  27. TA-GATES: An Encoding Scheme for Neural Network ArchitecturesAuthor 9 of 9NeurIPS 2022 · 19 citations
  28. Evaluating Efficient Performance Estimators of Neural ArchitecturesAuthor 7 of 7NeurIPS 2021 · 99 citations
  29. GE-SpMM: general-purpose sparse matrix-matrix multiplication on GPUs for graph neural networksAuthor 3 of 4SC 2020 · 130 citations
  30. Learning to Adapt to Evolving DomainsAuthor 4 of 4NeurIPS 2020 · 63 citations

Showing the 30 most recent indexed papers.

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