高效训练论文趋势 2023–2026

Scaling law、MoE、蒸馏与训练效率

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 1,765 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

高效训练论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
2023112678384346
2024207812280300
202519012611882516
2026177148169109603

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

全部高效训练论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 499 篇，第 8 / 10 页

  1. NeurIPS 2025SubTrack++ : Gradient Subspace Tracking for Scalable LLM Training
  2. NeurIPS 2025SUMO: Subspace-Aware Moment-Orthogonalization for Accelerating Memory-Efficient LLM Training
  3. NeurIPS 2025Superposition Yields Robust Neural Scaling
  4. NeurIPS 2025SymMaP: Improving Computational Efficiency in Linear Solvers through Symbolic Preconditioning
  5. NeurIPS 2025Synergy Between the Strong and the Weak: Spiking Neural Networks are Inherently Self-Distillers
  6. NeurIPS 2025Teaching Transformers to Solve Combinatorial Problems through Efficient Trial & Error
  7. NeurIPS 2025Technical Debt in In-Context Learning: Diminishing Efficiency in Long Context
  8. NeurIPS 2025The Omni-Expert: A Computationally Efficient Approach to Achieve a Mixture of Experts in a Single Expert Model
  9. NeurIPS 2025Tiled Flash Linear Attention: More Efficient Linear RNN and xLSTM Kernels
  10. NeurIPS 2025Towards Interpretable and Efficient Attention: Compressing All by Contracting a Few
  11. NeurIPS 2025TrajMamba: An Efficient and Semantic-rich Vehicle Trajectory Pre-training Model
  12. NeurIPS 2025UMoE: Unifying Attention and FFN with Shared Experts
  13. NeurIPS 2025Unified Scaling Laws for Compressed Representations
  14. NeurIPS 2025V2V: Scaling Event-Based Vision through Efficient Video-to-Voxel Simulation
  15. NeurIPS 2025Venus-MAXWELL: Efficient Learning of Protein-Mutation Stability Landscapes using Protein Language Models
  16. NeurIPS 2025Zebra-Llama: Towards Extremely Efficient Hybrid Models
  17. NeurIPS 2025ZeroS: Zero‑Sum Linear Attention for Efficient Transformers
  18. NeurIPS 2024$\texttt{dattri}$: A Library for Efficient Data Attribution
  19. NeurIPS 2024A Label is Worth A Thousand Images in Dataset Distillation
  20. NeurIPS 2024AmoebaLLM: Constructing Any-Shape Large Language Models for Efficient and Instant Deployment
  21. NeurIPS 2024BAM! Just Like That: Simple and Efficient Parameter Upcycling for Mixture of Experts
  22. NeurIPS 2024DeltaDock: A Unified Framework for Accurate, Efficient, and Physically Reliable Molecular Docking
  23. NeurIPS 2024E2ENet: Dynamic Sparse Feature Fusion for Accurate and Efficient 3D Medical Image Segmentation
  24. NeurIPS 2024Efficient Convex Algorithms for Universal Kernel Learning
  25. NeurIPS 2024FedGMKD: An Efficient Prototype Federated Learning Framework through Knowledge Distillation and Discrepancy-Aware Aggregation
  26. NeurIPS 2024Genetic-guided GFlowNets for Sample Efficient Molecular Optimization
  27. NeurIPS 2024LLMCBench: Benchmarking Large Language Model Compression for Efficient Deployment
  28. NeurIPS 2024LSH-MoE: Communication-efficient MoE Training via Locality-Sensitive Hashing
  29. NeurIPS 2024Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length
  30. NeurIPS 2024Provable and Efficient Dataset Distillation for Kernel Ridge Regression
  31. NeurIPS 2024Rough Transformers: Lightweight and Continuous Time Series Modelling through Signature Patching
  32. NeurIPS 2024Scaling Laws with Vocabulary: Larger Models Deserve Larger Vocabularies
  33. NeurIPS 2024Searching for Efficient Linear Layers over a Continuous Space of Structured Matrices
  34. NeurIPS 2024SLTrain: a sparse plus low rank approach for parameter and memory efficient pretraining
  35. NeurIPS 2024SpaFL: Communication-Efficient Federated Learning With Sparse Models And Low Computational Overhead
  36. NeurIPS 2024TorchOpt: An Efficient Library for Differentiable Optimization
  37. NeurIPS 2024V-PETL Bench: A Unified Visual Parameter-Efficient Transfer Learning Benchmark
  38. NeurIPS 2023A Computation and Communication Efficient Method for Distributed Nonconvex Problems in the Partial Participation Setting
  39. NeurIPS 2023A Hierarchical Training Paradigm for Antibody Structure-sequence Co-design
  40. NeurIPS 2023A Smooth Binary Mechanism for Efficient Private Continual Observation
  41. NeurIPS 2023Accelerated Training via Incrementally Growing Neural Networks using Variance Transfer and Learning Rate Adaptation
  42. NeurIPS 2023Actively Testing Your Model While It Learns: Realizing Label-Efficient Learning in Practice
  43. NeurIPS 2023Adversarial Training from Mean Field Perspective
  44. NeurIPS 2023Alternating Updates for Efficient Transformers
  45. NeurIPS 2023Amortized Reparametrization: Efficient and Scalable Variational Inference for Latent SDEs
  46. NeurIPS 2023An Efficient and Robust Framework for Approximate Nearest Neighbor Search with Attribute Constraint
  47. NeurIPS 2023An Efficient End-to-End Training Approach for Zero-Shot Human-AI Coordination
  48. NeurIPS 2023An Empirical Study Towards Prompt-Tuning for Graph Contrastive Pre-Training in Recommendations
  49. NeurIPS 2023An Inverse Scaling Law for CLIP Training
  50. NeurIPS 2023ATMAN: Understanding Transformer Predictions Through Memory Efficient Attention Manipulation

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查高效训练在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "efficient-training"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

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。

「效率与系统」下的其他主题