稀疏加速论文趋势 2023–2026

主要结论

• 稀疏加速论文从 2023 年的 54 篇增长到 2026 年的 89 篇，三年间扩大到约 1.6 倍。
• NeurIPS 是发稀疏加速论文最多的会议，2023–2026 年共 80 篇，占四个会议总量的 31%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 13 篇。
• 在「硬件与编译」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的端侧推理是 4.0 倍，稀疏加速是 1.6 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026AC/DC on a Budget -- Alternating Sparse Phases
• ICML 2026An Evidential Route to Asymptotic Bayes Optimality under Sparsity
• ICLR 2026Hilbert-Guided Sparse Local Attention
• CVPR 2026Improving Sparse Autoencoder with Dynamic Attention
• NeurIPS 2025Bilevel Network Learning via Hierarchically Structured Sparsity
• ICML 2025COGNATE: Acceleration of Sparse Tensor Programs on Emerging Hardware using Transfer Learning
• ICLR 2025Beyond Worst-Case Dimensionality Reduction for Sparse Vectors
• CVPR 2025ICP: Immediate Compensation Pruning for Mid-to-high Sparsity
• NeurIPS 2024ELSA: Exploiting Layer-wise N:M Sparsity for Vision Transformer Acceleration
• ICML 2024Accelerating Transformer Pre-training with 2:4 Sparsity
• ICLR 2024Adaptive Sharpness-Aware Pruning for Robust Sparse Networks
• CVPR 2024Masked Spatial Propagation Network for Sparsity-Adaptive Depth Refinement
• NeurIPS 2023Dynamic Sparsity Is Channel-Level Sparsity Learner
• ICML 2023Conformalization of Sparse Generalized Linear Models
• ICLR 2023Bit-Pruning: A Sparse Multiplication-Less Dot-Product
• CVPR 2023Adaptive Channel Sparsity for Federated Learning Under System Heterogeneity
• NeurIPS 2026Beyond Masked Sparsity: SNACK Enables Truly Sparse Neural Networks on GPU
• ICML 2026FlashSketch: Sketch-Kernel Co-Design for Fast Sparse Sketching on GPUs
• ICLR 2026Identifiability Challenges in Sparse Linear Ordinary Differential Equations
• CVPR 2026Revisiting Pose Sensitivity in Splat-based Computed Tomography under Sparse-view Reconstruction

全部稀疏加速论文

共 53 篇，第 1 / 2 页

1. CVPR 2024Masked Spatial Propagation Network for Sparsity-Adaptive Depth Refinement
2. CVPR 2024MaxQ: Multi-Axis Query for N:M Sparsity Network
3. CVPR 2024Towards More Accurate Diffusion Model Acceleration with A Timestep Tuner
4. CVPR 2024Transferable Structural Sparse Adversarial Attack Via Exact Group Sparsity Training
5. CVPR 2024UniPTS: A Unified Framework for Proficient Post-Training Sparsity
6. ICLR 2024Adaptive Sharpness-Aware Pruning for Robust Sparse Networks
7. ICLR 2024Balancing Act: Constraining Disparate Impact in Sparse Models
8. ICLR 2024DEEP NEURAL NETWORK INITIALIZATION WITH SPARSITY INDUCING ACTIVATIONS
9. ICLR 2024Dynamic Sparse No Training: Training-Free Fine-tuning for Sparse LLMs
10. ICLR 2024Dynamic Sparse Training with Structured Sparsity
11. ICLR 2024ReLU Strikes Back: Exploiting Activation Sparsity in Large Language Models
12. ICLR 2024SAS: Structured Activation Sparsification
13. ICLR 2024Sparsistency for inverse optimal transport
14. ICLR 2024SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression
15. ICLR 2024Training Bayesian Neural Networks with Sparse Subspace Variational Inference
16. ICML 2024Accelerating Transformer Pre-training with 2:4 Sparsity
17. ICML 2024Better Locally Private Sparse Estimation Given Multiple Samples Per User
18. ICML 2024Block Acceleration Without Momentum: On Optimal Stepsizes of Block Gradient Descent for Least-Squares
19. ICML 2024Distributional Bellman Operators over Mean Embeddings
20. ICML 2024Diversified Batch Selection for Training Acceleration
21. ICML 2024Enhancing Adversarial Robustness in SNNs with Sparse Gradients
22. ICML 2024Exploring the Benefit of Activation Sparsity in Pre-training
23. ICML 2024High-dimensional Linear Bandits with Knapsacks
24. ICML 2024Outlier Weighed Layerwise Sparsity (OWL): A Missing Secret Sauce for Pruning LLMs to High Sparsity
25. ICML 2024SPADE: Sparsity-Guided Debugging for Deep Neural Networks
26. ICML 2024Sparse Cocktail: Every Sparse Pattern Every Sparse Ratio All At Once
27. ICML 2024Sparse Dimensionality Reduction Revisited
28. ICML 2024Sparse is Enough in Fine-tuning Pre-trained Large Language Models
29. ICML 2024Sparse-IFT: Sparse Iso-FLOP Transformations for Maximizing Training Efficiency
30. ICML 2024Sparser, Better, Deeper, Stronger: Improving Static Sparse Training with Exact Orthogonal Initialization
31. ICML 2024Straight-Through Meets Sparse Recovery: the Support Exploration Algorithm
32. ICML 2024Submodular framework for structured-sparse optimal transport
33. ICML 2024Transformers Provably Learn Sparse Token Selection While Fully-Connected Nets Cannot
34. NeurIPS 2024Adaptive Layer Sparsity for Large Language Models via Activation Correlation Assessment
35. NeurIPS 2024AsyncDiff: Parallelizing Diffusion Models by Asynchronous Denoising
36. NeurIPS 2024Block Sparse Bayesian Learning: A Diversified Scheme
37. NeurIPS 2024CRONOS: Enhancing Deep Learning with Scalable GPU Accelerated Convex Neural Networks
38. NeurIPS 2024Discovering Sparsity Allocation for Layer-wise Pruning of Large Language Models
39. NeurIPS 2024ELSA: Exploiting Layer-wise N:M Sparsity for Vision Transformer Acceleration
40. NeurIPS 2024Fast Iterative Hard Thresholding Methods with Pruning Gradient Computations
41. NeurIPS 2024How Sparse Can We Prune A Deep Network: A Fundamental Limit Perspective
42. NeurIPS 2024Improving Decision Sparsity
43. NeurIPS 2024Learn To be Efficient: Build Structured Sparsity in Large Language Models
44. NeurIPS 2024Learning from Highly Sparse Spatio-temporal Data
45. NeurIPS 2024MaskLLM: Learnable Semi-Structured Sparsity for Large Language Models
46. NeurIPS 2024Navigating Extremes: Dynamic Sparsity in Large Output Spaces
47. NeurIPS 2024Nesterov acceleration despite very noisy gradients
48. NeurIPS 2024Numerically Stable Sparse Gaussian Processes via Minimum Separation using Cover Trees
49. NeurIPS 2024On the Sparsity of the Strong Lottery Ticket Hypothesis
50. NeurIPS 2024S-STE: Continuous Pruning Function for Efficient 2:4 Sparse Pre-training

常见问题

2023–2026 年四大会议一共有多少篇稀疏加速论文？

共 260 篇，按年份分别是 2023 年 54 篇、2024 年 53 篇、2025 年 64 篇、2026 年 89 篇。

NeurIPS 2026 有多少篇稀疏加速论文？

NeurIPS 2026 有 29 篇稀疏加速论文，上一届是 21 篇。

ICLR 2026 有多少篇稀疏加速论文？

ICLR 2026 有 24 篇稀疏加速论文，上一届是 16 篇。

ICML 2026 有多少篇稀疏加速论文？

ICML 2026 有 24 篇稀疏加速论文，上一届是 22 篇。

CVPR 2026 有多少篇稀疏加速论文？

CVPR 2026 有 12 篇稀疏加速论文，上一届是 5 篇。

哪个会议的稀疏加速论文最多？

NeurIPS，2023–2026 年共 80 篇。