Lune

Profiles2020-2025

Liwei Wang

Peking University

20Papers
14Last author
41Unique co-authors
0First author
0Solo

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

Recent papers20

  1. Theoretical Benefit and Limitation of Diffusion Language ModelAuthor 5 of 6NeurIPS 2025 · 52 citations
  2. UniSite: The First Cross-Structure Dataset and Learning Framework for End-to-End Ligand Binding Site DetectionAuthor 4 of 4NeurIPS 2025 · 7 citations
  3. DeformCL: Learning Deformable Centerline Representation for Vessel Extraction in 3D Medical ImageAuthor 7 of 7CVPR 2025
  4. Homomorphism Expressivity of Spectral Invariant Graph Neural NetworksAuthor 5 of 5ICLR 2025
  5. Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale PredictionAuthor 5 of 5NeurIPS 2024 · 1,199 citations
  6. Beyond Weisfeiler-Lehman: A Quantitative Framework for GNN ExpressivenessAuthor 6 of 6ICLR 2024 · 59 citations
  7. Do Efficient Transformers Really Save Computation?Author 9 of 9ICML 2024 · 29 citations
  8. GraphMorph: Tubular Structure Extraction by Morphing Predicted GraphsAuthor 4 of 4NeurIPS 2024 · 4 citations
  9. Bridging Geometric States via Geometric Diffusion BridgeAuthor 6 of 6NeurIPS 2024 · 3 citations
  10. Towards Revealing the Mystery behind Chain of Thought: A Theoretical PerspectiveAuthor 6 of 6NeurIPS 2023 · 470 citations
  11. Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked ModelingAuthor 5 of 6ICLR 2023 · 37 citations
  12. Denoising Masked Autoencoders Help Robust ClassificationAuthor 5 of 6ICLR 2023 · 5 citations
  13. One Transformer Can Understand Both 2D & 3D Molecular DataAuthor 6 of 7ICLR 2023 · 4 citations
  14. Rethinking Lipschitz Neural Networks and Certified Robustness: A Boolean Function PerspectiveAuthor 4 of 4NeurIPS 2022 · 88 citations
  15. Your Transformer May Not be as Powerful as You ExpectAuthor 5 of 6NeurIPS 2022 · 69 citations
  16. Is L2L^2 Physics Informed Loss Always Suitable for Training Physics Informed Neural Network?Author 4 of 4NeurIPS 2022 · 36 citations
  17. Stable, Fast and Accurate: Kernelized Attention with Relative Positional EncodingAuthor 8 of 9NeurIPS 2021 · 66 citations
  18. Augmented Partial Mutual Learning with Frame Masking for Video CaptioningAuthor 3 of 3AAAI 2021 · 25 citations
  19. Semi-Supervised Learning for Video CaptioningAuthor 3 of 3EMNLP Findings 2020 · 4 citations
  20. FocalMix: Semi-Supervised Learning for 3D Medical Image DetectionAuthor 4 of 4CVPR 2020
Dusk over the sea between two cliffs drawn in fine vertical lines