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 516 篇，第 7 / 11 页

301. ICML 2025Scalable Model Merging with Progressive Layer-wise Distillation
302. ICML 2025Scaling Collapse Reveals Universal Dynamics in Compute-Optimally Trained Neural Networks
303. ICML 2025Scaling Inference-Efficient Language Models
304. ICML 2025Scaling Laws for Floating–Point Quantization Training
305. ICML 2025Scaling Probabilistic Circuits via Monarch Matrices
306. ICML 2025SDP-CROWN: Efficient Bound Propagation for Neural Network Verification with Tightness of Semidefinite Programming
307. ICML 2025SKIM: Any-bit Quantization Pushing The Limits of Post-Training Quantization
308. ICML 2025Sorbet: A Neuromorphic Hardware-Compatible Transformer-Based Spiking Language Model
309. ICML 2025Sparse Spectral Training and Inference on Euclidean and Hyperbolic Neural Networks
310. ICML 2025SpikeVideoFormer: An Efficient Spike-Driven Video Transformer with Hamming Attention and $\mathcal{O}(T)$ Complexity
311. ICML 2025SpikF: Spiking Fourier Network for Efficient Long-term Prediction
312. ICML 2025Star Attention: Efficient LLM Inference over Long Sequences
313. ICML 2025SWAN: SGD with Normalization and Whitening Enables Stateless LLM Training
314. ICML 2025TabFlex: Scaling Tabular Learning to Millions with Linear Attention
315. ICML 2025Temporal Query Network for Efficient Multivariate Time Series Forecasting
316. ICML 2025Tensor Decomposition Based Memory-Efficient Incremental Learning
317. ICML 2025The Hidden Joules: Evaluating the Energy Consumption of Vision Backbones for Progress Towards More Efficient Model Inference
318. ICML 2025The Limits of Tractable Marginalization
319. ICML 2025The Panaceas for Improving Low-Rank Decomposition in Communication-Efficient Federated Learning
320. ICML 2025TimeBase: The Power of Minimalism in Efficient Long-term Time Series Forecasting
321. ICML 2025TimePro: Efficient Multivariate Long-term Time Series Forecasting with Variable- and Time-Aware Hyper-state
322. ICML 2025Training Deep Learning Models with Norm-Constrained LMOs
323. ICML 2025UncertainSAM: Fast and Efficient Uncertainty Quantification of the Segment Anything Model
324. ICML 2025Unifews: You Need Fewer Operations for Efficient Graph Neural Networks
325. ICML 2025XAttention: Block Sparse Attention with Antidiagonal Scoring
326. ICML 2025xLSTM 7B: A Recurrent LLM for Fast and Efficient Inference
327. NeurIPS 2025$\texttt{BetaConform}$: Efficient MAP Estimation of LLM Ensemble Judgment Performance with Prior Transfer
328. NeurIPS 20254DGCPro: Efficient Hierarchical 4D Gaussian Compression for Progressive Volumetric Video Streaming
329. NeurIPS 2025A Stable Whitening Optimizer for Efficient Neural Network Training
330. NeurIPS 2025A Token is Worth over 1,000 Tokens: Efficient Knowledge Distillation through Low-Rank Clone
331. NeurIPS 2025A Unified Framework for Provably Efficient Algorithms to Estimate Shapley Values
332. NeurIPS 2025Accurate and Efficient Low-Rank Model Merging in Core Space
333. NeurIPS 2025Accurate KV Cache Eviction via Anchor Direction Projection for Efficient LLM Inference
334. NeurIPS 2025Active Measurement: Efficient Estimation at Scale
335. NeurIPS 2025Active Seriation: Efficient Ordering Recovery with Statistical Guarantees
336. NeurIPS 2025Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference
337. NeurIPS 2025AdaSPEC: Selective Knowledge Distillation for Efficient Speculative Decoders
338. NeurIPS 2025All You Need is One: Capsule Prompt Tuning with a Single Vector
339. NeurIPS 2025AmorLIP: Efficient Language-Image Pretraining via Amortization
340. NeurIPS 2025An Efficient Local Search Approach for Polarized Community Discovery in Signed Networks
341. NeurIPS 2025Angles Don’t Lie: Unlocking Training‑Efficient RL Through the Model’s Own Signals
342. NeurIPS 2025Are Language Models Efficient Reasoners? A Perspective from Logic Programming
343. NeurIPS 2025Audits Under Resource, Data, and Access Constraints: Scaling Laws For Less Discriminatory Alternatives
344. NeurIPS 2025Bilevel ZOFO: Efficient LLM Fine-Tuning and Meta-Training
345. NeurIPS 2025Breaking the Compression Ceiling: Data-Free Pipeline for Ultra-Efficient Delta Compression
346. NeurIPS 2025CAT: Circular-Convolutional Attention for Sub-Quadratic Transformers
347. NeurIPS 2025Causal Spatio-Temporal Prediction: An Effective and Efficient Multi-Modal Approach
348. NeurIPS 2025ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference
349. NeurIPS 2025COALA: Numerically Stable and Efficient Framework for Context-Aware Low-Rank Approximation
350. NeurIPS 2025CodeGEMM: A Codebook-Centric Approach to Efficient GEMM in Quantized LLMs

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。