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

主要结论

• 深度学习 · representation learning论文从 2023 年的 259 篇降到 2026 年的 214 篇，热度在回落。
• NeurIPS 是发深度学习 · representation learning论文最多的会议，2023–2026 年共 279 篇，占四个会议总量的 35%。
• 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 11 篇增至 2026 年的 28 篇。
• 在「深度学习」的 11 个主题里，它的 2026 年论文数排第 7；同组增长最快的扩散模型是 4.1 倍，深度学习 · representation learning是 0.8 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 2026Active Learning as Nullspace Regulation: A Spectral Representation Perspective
• ICML 2026Addressing Instrument-Outcome Confounding in Mendelian Randomization through Representation Learning
• ICLR 2026Boosting Open Set Recognition Performance through Modulated Representation Learning
• CVPR 2026CanonCGT Reference-Based Color Grading via Canonical Pivot Representation
• NeurIPS 2025Causality-Induced Positional Encoding for Transformer-Based Representation Learning of Non-Sequential Features
• ICML 2025Banyan: Improved Representation Learning with Explicit Structure
• ICLR 2025Edge-aware Image Smoothing with Relative Wavelet Domain Representation
• CVPR 20253D-SLNR: A Super Lightweight Neural Representation for Large-scale 3D Mapping
• NeurIPS 2024GV-Rep: A Large-Scale Dataset for Genetic Variant Representation Learning
• ICML 2024Data-free Neural Representation Compression with Riemannian Neural Dynamics
• ICLR 2024$\texttt{NAISR}$: A 3D Neural Additive Model for Interpretable Shape Representation
• CVPR 2024Disentangled Prompt Representation for Domain Generalization
• NeurIPS 2023[Re] CrossWalk: Fairness-enhanced Node Representation Learning
• ICML 2023A theory of representation learning gives a deep generalisation of kernel methods
• ICLR 2023Bispectral Neural Networks
• CVPR 2023AnyFlow: Arbitrary Scale Optical Flow With Implicit Neural Representation
• NeurIPS 2026Approximation in Contrastive Representation Learning
• ICML 2026Amortized Maximum Inner Product Search with Learned Support Functions
• ICLR 2026Decoupling Dynamical Richness from Representation Learning: Towards Practical Measurement
• CVPR 2026Enhancing Visual Representation with Textual Semantics Textual Semantics-Powered Prototypes for

全部深度学习 · representation learning论文

共 152 篇，第 3 / 4 页

101. NeurIPS 2025Anti-Aliased 2D Gaussian Splatting
102. NeurIPS 2025Auto-Compressing Networks
103. NeurIPS 2025Beyond Modality Collapse: Representation Blending for Multimodal Dataset Distillation
104. NeurIPS 2025Causality-Induced Positional Encoding for Transformer-Based Representation Learning of Non-Sequential Features
105. NeurIPS 2025Conditional Representation Learning for Customized Tasks
106. NeurIPS 2025Connecting Neural Models Latent Geometries with Relative Geodesic Representations
107. NeurIPS 2025DISCO: Disentangled Communication Steering for Large Language Models
108. NeurIPS 2025Disentangled Cross-Modal Representation Learning with Enhanced Mutual Supervision
109. NeurIPS 2025Disentangled Representation Learning via Modular Compositional Bias
110. NeurIPS 2025Equivariance by Contrast: Identifiable Equivariant Embeddings from Unlabeled Finite Group Actions
111. NeurIPS 2025EventMG: Efficient Multilevel Mamba-Graph Learning for Spatiotemporal Event Representation
112. NeurIPS 2025FORLA: Federated Object-centric Representation Learning with Slot Attention
113. NeurIPS 2025GAMMA: Gated Multi-hop Message Passing for Homophily-Agnostic Node Representation in GNNs
114. NeurIPS 2025Global Minimizers of Sigmoid Contrastive Loss
115. NeurIPS 2025Grids Often Outperform Implicit Neural Representation at Compressing Dense Signals
116. NeurIPS 2025H3D-DGS: Exploring Heterogeneous 3D Motion Representation for Deformable 3D Gaussian Splatting
117. NeurIPS 2025How Many Tokens Do 3D Point Cloud Transformer Architectures Really Need?
118. NeurIPS 2025Kernel-based Equalized Odds: A Quantification of Accuracy-Fairness Trade-off in Fair Representation Learning
119. NeurIPS 2025Learning Differential Pyramid Representation for Tone Mapping
120. NeurIPS 2025Learning Generalizable Shape Completion with SIM(3) Equivariance
121. NeurIPS 2025Learning Without Augmenting: Unsupervised Time Series Representation Learning via Frame Projections
122. NeurIPS 2025LLM Interpretability with Identifiable Temporal-Instantaneous Representation
123. NeurIPS 2025MaxSup: Overcoming Representation Collapse in Label Smoothing
124. NeurIPS 2025Measure-Theoretic Anti-Causal Representation Learning
125. NeurIPS 2025Mitigating Spurious Features in Contrastive Learning with Spectral Regularization
126. NeurIPS 2025More of the Same: Persistent Representational Harms Under Increased Representation
127. NeurIPS 2025NegoCollab: A Common Representation Negotiation Approach for Heterogeneous Collaborative Perception
128. NeurIPS 2025On the Value of Cross-Modal Misalignment in Multimodal Representation Learning
129. NeurIPS 2025Permutation Equivariant Neural Controlled Differential Equations for Dynamic Graph Representation Learning
130. NeurIPS 2025PhysioWave: A Multi-Scale Wavelet-Transformer for Physiological Signal Representation
131. NeurIPS 2025Point or Line? Using Line-based Representation for Panoptic Symbol Spotting in CAD Drawings
132. NeurIPS 2025PoLAR: Polar-Decomposed Low-Rank Adapter Representation
133. NeurIPS 2025Provable Sample-Efficient Transfer Learning Conditional Diffusion Models via Representation Learning
134. NeurIPS 2025Random Forest Autoencoders for Guided Representation Learning
135. NeurIPS 2025REPA Works Until It Doesn’t: Early-Stopped, Holistic Alignment Supercharges Diffusion Training
136. NeurIPS 2025Reward-oriented Causal Representation Learning
137. NeurIPS 2025S-Crescendo: A Nested Transformer Weaving Framework for Scalable Nonlinear System in S-Domain Representation
138. NeurIPS 2025Scaling Language-centric Omnimodal Representation Learning
139. NeurIPS 2025Symmetry-Preserving Conformer Ensemble Networks for Molecular Representation Learning
140. NeurIPS 2025Temporal Representation Alignment: Successor Features Enable Emergent Compositionality in Robot Instruction Following
141. NeurIPS 2025The Indra Representation Hypothesis
142. NeurIPS 2025Theory-Driven Label-Specific Representation for Incomplete Multi-View Multi-Label Learning
143. NeurIPS 2025Token Bottleneck: One Token to Remember Dynamics
144. NeurIPS 2025TopER: Topological Embeddings in Graph Representation Learning
145. NeurIPS 2025Toward Artificial Palpation: Representation Learning of Touch on Soft Bodies
146. NeurIPS 2025Towards Implicit Aggregation: Robust Image Representation for Place Recognition in the Transformer Era
147. NeurIPS 2025Tracing the Representation Geometry of Language Models from Pretraining to Post-training
148. NeurIPS 2025U-REPA: Aligning Diffusion U-Nets to ViTs
149. NeurIPS 2025VA-GS: Enhancing the Geometric Representation of Gaussian Splatting via View Alignment
150. NeurIPS 2025VQToken: Neural Discrete Token Representation Learning for Extreme Token Reduction in Video Large Language Models

常见问题

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

共 802 篇，按年份分别是 2023 年 259 篇、2024 年 177 篇、2025 年 152 篇、2026 年 214 篇。

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

NeurIPS 2026 有 59 篇深度学习 · representation learning论文，上一届是 54 篇。

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

ICLR 2026 有 39 篇深度学习 · representation learning论文，上一届是 37 篇。

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

ICML 2026 有 74 篇深度学习 · representation learning论文，上一届是 35 篇。

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

CVPR 2026 有 42 篇深度学习 · representation learning论文，上一届是 26 篇。

哪个会议的深度学习 · representation learning论文最多？

NeurIPS，2023–2026 年共 279 篇。