Lune

Profiles2022-2026

Jianfei Chen

Tsinghua university

27Papers
9Last author
15Unique co-authors
0First author
0Solo

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

Recent papers27

  1. TetraJet-v2: Accurate NVFP4 Training for Large Language Models with Oscillation Suppression and Outlier ControlAuthor 8 of 8ICML 2026 · 11 citations
  2. Achieving low-bit Muon through subspace preservation and grid quantizationAuthor 6 of 7ICLR 2026
  3. Pruning Large Language Models with Semi-Structural Adaptive Sparse TrainingAuthor 5 of 5AAAI 2025 · 25 citations
  4. Task-Specific Zero-Shot Quantization-Aware Training for Object DetectionAuthor 5 of 5ICCV 2025 · 2 citations
  5. Diffusion Bridge Implicit ModelsAuthor 3 of 5ICLR 2025
  6. Elucidating the Preconditioning in Consistency DistillationAuthor 3 of 5ICLR 2025
  7. FrameBridge: Improving Image-to-Video Generation with Bridge ModelsAuthor 6 of 6ICML 2025
  8. Maximum Redundancy Pruning: A Principle-Driven Layerwise Sparsity Allocation for LLMsAuthor 4 of 5ACM MM 2025
  9. On the Optimization and Generalization of Two-layer Transformers with Sign Gradient DescentAuthor 7 of 7ICLR 2025
  10. Oscillation-Reduced MXFP4 Training for Vision TransformersAuthor 4 of 4ICML 2025
  11. ReMoE: Fully Differentiable Mixture-of-Experts with ReLU RoutingAuthor 3 of 3ICLR 2025
  12. SageAttention2: Efficient Attention with Thorough Outlier Smoothing and Per-thread INT4 QuantizationAuthor 6 of 6ICML 2025
  13. Jetfire: Efficient and Accurate Transformer Pretraining with INT8 Data Flow and Per-Block QuantizationAuthor 5 of 6ICML 2024 · 35 citations
  14. Consistency Diffusion Bridge ModelsAuthor 3 of 5NeurIPS 2024 · 31 citations
  15. Accelerating Transformer Pre-training with 2: 4 SparsityAuthor 4 of 5ICML 2024 · 19 citations
  16. S-STE: Continuous Pruning Function for Efficient 2: 4 Sparse Pre-trainingAuthor 3 of 3NeurIPS 2024 · 14 citations
  17. C-GAIL: Stabilizing Generative Adversarial Imitation Learning with Control TheoryAuthor 4 of 5NeurIPS 2024 · 13 citations
  18. Efficient Backpropagation with Variance Controlled Adaptive SamplingAuthor 2 of 3ICLR 2024 · 5 citations
  19. DPM-Solver-v3: Improved Diffusion ODE Solver with Empirical Model StatisticsAuthor 3 of 4NeurIPS 2023 · 204 citations
  20. Training Transformers with 4-bit IntegersAuthor 3 of 4NeurIPS 2023 · 96 citations
  21. Memory Efficient Optimizers with 4-bit StatesAuthor 2 of 3NeurIPS 2023 · 72 citations
  22. Stabilizing GANs' Training with Brownian Motion ControllerAuthor 3 of 4ICML 2023 · 5 citations
  23. Preserving Pre-trained Features Helps Calibrate Fine-tuned Language ModelsAuthor 2 of 3ICLR 2023 · 1 citation
  24. DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 StepsAuthor 4 of 6NeurIPS 2022 · 2,653 citations
  25. Maximum Likelihood Training for Score-based Diffusion ODEs by High Order Denoising Score MatchingAuthor 4 of 6ICML 2022 · 109 citations
  26. GACT: Activation Compressed Training for Generic Network ArchitecturesAuthor 7 of 12ICML 2022 · 43 citations
  27. Fast Lossless Neural Compression with Integer-Only Discrete FlowsAuthor 2 of 5ICML 2022 · 8 citations
Dusk over the sea between two cliffs drawn in fine vertical lines