深度学习 · transfer learning论文趋势 2023–2026

主要结论

• 深度学习 · transfer learning论文从 2023 年的 56 篇降到 2026 年的 54 篇，热度在回落。
• NeurIPS 是发深度学习 · transfer learning论文最多的会议，2023–2026 年共 71 篇，占四个会议总量的 36%。
• 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 2 篇增至 2026 年的 5 篇。
• 在「深度学习」的 11 个主题里，它的 2026 年论文数排第 11；同组增长最快的扩散模型是 4.1 倍，深度学习 · transfer learning是 1.0 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 2026Emergent Misalignment as Data-Mediated Transfer
• ICML 2026Transfer Learning in High-dimensional Ising Models
• ICLR 2026TiTok: Transfer Token-level Knowledge via Contrastive Excess to Transplant LoRA
• CVPR 2026Unlocking Positive Transfer in Incrementally Learning Surgical Instruments A Self-reflection
• NeurIPS 2025Asymmetric Duos: Sidekicks Improve Uncertainty
• ICML 2025An Improved Clique-Picking Algorithm for Counting Markov Equivalent DAGs via Super Cliques Transfer
• ICLR 2025GraphBridge: Towards Arbitrary Transfer Learning in GNNs
• CVPR 2025A4A: Adapter for Adapter Transfer via All-for-All Mapping for Cross-Architecture Models
• NeurIPS 2024Transfer Learning with Informative Priors: Simple Baselines Better than Previously Reported
• ICML 2024Adversarially Robust Hypothesis Transfer Learning
• ICLR 2024Geometrically Aligned Transfer Encoder for Inductive Transfer in Regression Tasks
• CVPR 2024Geometry Transfer for Stylizing Radiance Fields
• NeurIPS 2023Diffused Redundancy in Pre-trained Representations
• ICML 2023Hypothesis Transfer Learning with Surrogate Classification Losses: Generalization Bounds through Algorithmic Stability
• ICLR 2023Transfer Learning with Deep Tabular Models
• CVPR 2023A Data-Based Perspective on Transfer Learning
• NeurIPS 2026Learning Rate Transfer for Hybrid Transformer-SSM Architectures
• ICML 2026Transfer Learning in Nonparametric Regression with Deep ReLU Networks
• ICLR 2026Detect, Decide, Unlearn: A Transfer-Aware Framework for Continual Learning
• CVPR 2026Nonlinear Color Transfer via Learnable Bezier Flows

全部深度学习 · transfer learning论文

共 43 篇，第 1 / 1 页

1. CVPR 2025A4A: Adapter for Adapter Transfer via All-for-All Mapping for Cross-Architecture Models
2. CVPR 2025Attention Distillation: A Unified Approach to Visual Characteristics Transfer
3. CVPR 2025HistoFS: Non-IID Histopathologic Whole Slide Image Classification via Federated Style Transfer with RoI-Preserving
4. CVPR 2025HSI: A Holistic Style Injector for Arbitrary Style Transfer
5. CVPR 2025ManipTrans: Efficient Dexterous Bimanual Manipulation Transfer via Residual Learning
6. CVPR 2025OmniStyle: Filtering High Quality Style Transfer Data at Scale
7. CVPR 2025StyleSSP: Sampling StartPoint Enhancement for Training-free Diffusion-based Method for Style Transfer
8. CVPR 2025Video Motion Transfer with Diffusion Transformers
9. ICLR 2025A Transfer Attack to Image Watermarks
10. ICLR 2025A transfer learning framework for weak to strong generalization
11. ICLR 2025GraphBridge: Towards Arbitrary Transfer Learning in GNNs
12. ICLR 2025Less is More: Masking Elements in Image Condition Features Avoids Content Leakages in Style Transfer Diffusion Models
13. ICLR 2025Let Me Grok for You: Accelerating Grokking via Embedding Transfer from a Weaker Model
14. ICLR 2025Query-based Knowledge Transfer for Heterogeneous Learning Environments
15. ICLR 2025Release the Powers of Prompt Tuning: Cross-Modality Prompt Transfer
16. ICLR 2025Semantix: An Energy-guided Sampler for Semantic Style Transfer
17. ICLR 2025The Case for Cleaner Biosignals: High-fidelity Neural Compressor Enables Transfer from Cleaner iEEG to Noisier EEG
18. ICML 2025An Improved Clique-Picking Algorithm for Counting Markov Equivalent DAGs via Super Cliques Transfer
19. ICML 2025Do Multiple Instance Learning Models Transfer?
20. ICML 2025Features are fate: a theory of transfer learning in high-dimensional regression
21. ICML 2025Function Encoders: A Principled Approach to Transfer Learning in Hilbert Spaces
22. ICML 2025Function-Space Learning Rates
23. ICML 2025Lightspeed Geometric Dataset Distance via Sliced Optimal Transport
24. ICML 2025On the Provable Separation of Scales in Maximal Update Parameterization
25. ICML 2025Overtrained Language Models Are Harder to Fine-Tune
26. ICML 2025Predicting mutational effects on protein binding from folding energy
27. ICML 2025TLLC: Transfer Learning-based Label Completion for Crowdsourcing
28. NeurIPS 2025$\mu$PC: Scaling Predictive Coding to 100+ Layer Networks
29. NeurIPS 2025$\texttt{BetaConform}$: Efficient MAP Estimation of LLM Ensemble Judgment Performance with Prior Transfer
30. NeurIPS 2025Asymmetric Duos: Sidekicks Improve Uncertainty
31. NeurIPS 2025Co-Regularization Enhances Knowledge Transfer in High Dimensions
32. NeurIPS 2025Don't be lazy: CompleteP enables compute-efficient deep transformers
33. NeurIPS 2025Extrapolation by Association: Length Generalization Transfer In Transformers
34. NeurIPS 2025How Classifier Features Transfer to Downstream: An Asymptotic Analysis in a Two-Layer Model
35. NeurIPS 2025One for All: Universal Topological Primitive Transfer for Graph Structure Learning
36. NeurIPS 2025Order-Level Attention Similarity Across Language Models: A Latent Commonality
37. NeurIPS 2025Pairwise Optimal Transports for Training All-to-All Flow-Based Condition Transfer Model
38. NeurIPS 2025Provable Sample-Efficient Transfer Learning Conditional Diffusion Models via Representation Learning
39. NeurIPS 2025Transfer Faster, Price Smarter: Minimax Dynamic Pricing under Cross-Market Preference Shift
40. NeurIPS 2025Transfer Learning for Benign Overfitting in High-Dimensional Linear Regression
41. NeurIPS 2025Transfer Learning on Edge Connecting Probability Estimation Under Graphon Model
42. NeurIPS 2025Universal Cross-Tokenizer Distillation via Approximate Likelihood Matching
43. NeurIPS 2025Wasserstein Transfer Learning

常见问题

2023–2026 年四大会议一共有多少篇深度学习 · transfer learning论文？

共 197 篇，按年份分别是 2023 年 56 篇、2024 年 44 篇、2025 年 43 篇、2026 年 54 篇。

NeurIPS 2026 有多少篇深度学习 · transfer learning论文？

NeurIPS 2026 有 18 篇深度学习 · transfer learning论文，上一届是 16 篇。

ICLR 2026 有多少篇深度学习 · transfer learning论文？

ICLR 2026 有 9 篇深度学习 · transfer learning论文，上一届是 9 篇。

ICML 2026 有多少篇深度学习 · transfer learning论文？

ICML 2026 有 13 篇深度学习 · transfer learning论文，上一届是 10 篇。

CVPR 2026 有多少篇深度学习 · transfer learning论文？

CVPR 2026 有 14 篇深度学习 · transfer learning论文，上一届是 8 篇。

哪个会议的深度学习 · transfer learning论文最多？

NeurIPS，2023–2026 年共 71 篇。