高效训练论文趋势 2023–2026

主要结论

• 高效训练论文从 2023 年的 346 篇增长到 2026 年的 603 篇，三年间扩大到约 1.7 倍。
• NeurIPS 是发高效训练论文最多的会议，2023–2026 年共 499 篇，占四个会议总量的 28%。
• 交叉方向里涨得最快的是深度学习 · large language models：同时属于两个主题的论文从 2023 年的 6 篇增至 2026 年的 50 篇。
• 在「效率与系统」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的硬件协同是 4.4 倍，高效训练是 1.7 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 2026Adaptive Residual Quantization for Memory-Efficient Temporal Action Segmentation
• ICML 2026A Geometry-Aware Efficient Algorithm for Compositional Entropic Risk Minimization
• ICLR 2026An Efficient SE(p)-Invariant Transport Metric Driven by Polar Transport Discrepancy-based Representation
• CVPR 2026A Bit is All You Need Efficient Video Capture via
• NeurIPS 20254DGCPro: Efficient Hierarchical 4D Gaussian Compression for Progressive Volumetric Video Streaming
• ICML 2025A Sample Efficient Conditional Independence Test in the Presence of Discretization
• ICLR 2025Breaking Class Barriers: Efficient Dataset Distillation via Inter-Class Feature Compensator
• CVPR 2025Accurate Scene Text Recognition with Efficient Model Scaling and Cloze Self-Distillation
• NeurIPS 2024$\texttt{dattri}$: A Library for Efficient Data Attribution
• ICML 2024A fast algorithm to simulate nonlinear resistive networks
• ICLR 2024Dictionary Contrastive Learning for Efficient Local Supervision without Auxiliary Networks
• CVPR 2024Deep-TROJ: An Inference Stage Trojan Insertion Algorithm through Efficient Weight Replacement Attack
• NeurIPS 2023Alternating Updates for Efficient Transformers
• ICML 2023Auxiliary Modality Learning with Generalized Curriculum Distillation
• ICLR 2023A Model or 603 Exemplars: Towards Memory-Efficient Class-Incremental Learning
• CVPR 2023A Rotation-Translation-Decoupled Solution for Robust and Efficient Visual-Inertial Initialization
• NeurIPS 2026Advanced Routing as Regularization Allocation for Efficient Diffusion Transformer Training
• ICML 2026Active Learning with Foundation Model Priors: Efficient Learning under Class Imbalance
• ICLR 2026AQER: A Scalable and Efficient Data Loader for Digital Quantum Computers
• CVPR 2026Adaptive Bayesian Early-Exit Networks for Efficient Non-Transferable Learning

全部高效训练论文

共 118 篇，第 2 / 3 页

51. ICML 2025FSL-SAGE: Accelerating Federated Split Learning via Smashed Activation Gradient Estimation
52. ICML 2025GPTAQ: Efficient Finetuning-Free Quantization for Asymmetric Calibration
53. ICML 2025H-Tuning: Toward Low-Cost and Efficient ECG-based Cardiovascular Disease Detection with Pre-Trained Models
54. ICML 2025HiRemate: Hierarchical Approach for Efficient Re-materialization of Neural Networks
55. ICML 2025How to Train Your Multi-Exit Model? Analyzing the Impact of Training Strategies
56. ICML 2025IL-SOAR : Imitation Learning with Soft Optimistic Actor cRitic
57. ICML 2025Improved and Oracle-Efficient Online $\ell_1$-Multicalibration
58. ICML 2025Improving Continual Learning Performance and Efficiency with Auxiliary Classifiers
59. ICML 2025Info-Coevolution: An Efficient Framework for Data Model Coevolution
60. ICML 2025Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient
61. ICML 2025KVTuner: Sensitivity-Aware Layer-Wise Mixed-Precision KV Cache Quantization for Efficient and Nearly Lossless LLM Inference
62. ICML 2025LaCache: Ladder-Shaped KV Caching for Efficient Long-Context Modeling of Large Language Models
63. ICML 2025Learning from Loss Landscape: Generalizable Mixed-Precision Quantization via Adaptive Sharpness-Aware Gradient Aligning
64. ICML 2025Learning Soft Sparse Shapes for Efficient Time-Series Classification
65. ICML 2025LensLLM: Unveiling Fine-Tuning Dynamics for LLM Selection
66. ICML 2025Leveraging Sparsity for Sample-Efficient Preference Learning: A Theoretical Perspective
67. ICML 2025Memory Layers at Scale
68. ICML 2025MERGE$^3$: Efficient Evolutionary Merging on Consumer-grade GPUs
69. ICML 2025Mixture of Hidden-Dimensions: Not All Hidden-States’ Dimensions are Needed in Transformer
70. ICML 2025Model Steering: Learning with a Reference Model Improves Generalization Bounds and Scaling Laws
71. ICML 2025Nonparametric Modern Hopfield Models
72. ICML 2025Occult: Optimizing Collaborative Communications across Experts for Accelerated Parallel MoE Training and Inference
73. ICML 2025Olica: Efficient Structured Pruning of Large Language Models without Retraining
74. ICML 2025On Efficient Estimation of Distributional Treatment Effects under Covariate-Adaptive Randomization
75. ICML 2025On the Duality between Gradient Transformations and Adapters
76. ICML 2025On-the-Fly Adaptive Distillation of Transformer to Dual-State Linear Attention for Long-Context LLM Serving
77. ICML 2025One Leaf Reveals the Season: Occlusion-Based Contrastive Learning with Semantic-Aware Views for Efficient Visual Representation
78. ICML 2025OrthoRank: Token Selection via Sink Token Orthogonality for Efficient LLM inference
79. ICML 2025Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling
80. ICML 2025Parameter-Efficient Fine-Tuning of State Space Models
81. ICML 2025Position: Enough of Scaling LLMs! Lets Focus on Downscaling
82. ICML 2025Predictive Data Selection: The Data That Predicts Is the Data That Teaches
83. ICML 2025Primphormer: Efficient Graph Transformers with Primal Representations
84. ICML 2025ProofAug: Efficient Neural Theorem Proving via Fine-grained Proof Structure Analysis
85. ICML 2025QoS-Efficient Serving of Multiple Mixture-of-Expert LLMs Using Partial Runtime Reconfiguration
86. ICML 2025Reliable and Efficient Amortized Model-based Evaluation
87. ICML 2025Residual Matrix Transformers: Scaling the Size of the Residual Stream
88. ICML 2025Reward-Guided Speculative Decoding for Efficient LLM Reasoning
89. ICML 2025Robust Consensus Anchor Learning for Efficient Multi-view Subspace Clustering
90. ICML 2025RoSTE: An Efficient Quantization-Aware Supervised Fine-Tuning Approach for Large Language Models
91. ICML 2025SageAttention2: Efficient Attention with Thorough Outlier Smoothing and Per-thread INT4 Quantization
92. ICML 2025Sample Efficient Demonstration Selection for In-Context Learning
93. ICML 2025Scalable Model Merging with Progressive Layer-wise Distillation
94. ICML 2025Scaling Collapse Reveals Universal Dynamics in Compute-Optimally Trained Neural Networks
95. ICML 2025Scaling Inference-Efficient Language Models
96. ICML 2025Scaling Laws for Floating–Point Quantization Training
97. ICML 2025Scaling Probabilistic Circuits via Monarch Matrices
98. ICML 2025SDP-CROWN: Efficient Bound Propagation for Neural Network Verification with Tightness of Semidefinite Programming
99. ICML 2025SKIM: Any-bit Quantization Pushing The Limits of Post-Training Quantization
100. ICML 2025Sorbet: A Neuromorphic Hardware-Compatible Transformer-Based Spiking Language Model

常见问题

2023–2026 年四大会议一共有多少篇高效训练论文？

共 1,765 篇，按年份分别是 2023 年 346 篇、2024 年 300 篇、2025 年 516 篇、2026 年 603 篇。

NeurIPS 2026 有多少篇高效训练论文？

NeurIPS 2026 有 177 篇高效训练论文，上一届是 190 篇。

ICLR 2026 有多少篇高效训练论文？

ICLR 2026 有 148 篇高效训练论文，上一届是 126 篇。

ICML 2026 有多少篇高效训练论文？

ICML 2026 有 169 篇高效训练论文，上一届是 118 篇。

CVPR 2026 有多少篇高效训练论文？

CVPR 2026 有 109 篇高效训练论文，上一届是 82 篇。

哪个会议的高效训练论文最多？

NeurIPS，2023–2026 年共 499 篇。