域适应论文趋势 2023–2026

主要结论

• 域适应论文从 2023 年的 174 篇增长到 2026 年的 263 篇，三年间扩大到约 1.5 倍。
• CVPR 是发域适应论文最多的会议，2023–2026 年共 268 篇，占四个会议总量的 31%。
• 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 7 篇增至 2026 年的 38 篇。
• 在「元学习与迁移」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的基础模型适配是 3.8 倍，域适应是 1.5 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 2026A Dual-Domain Vision Transformer with Spectral Positional Bias
• ICML 2026Bayesian Meta-Learning with Expert Feedback for Task-Shift Adaptation through Causal Embeddings
• ICLR 2026Bayesian Evidence-Driven Prototype Evolution for Federated Domain Adaptation
• CVPR 2026Anatomical Domain Shifts Test-time Heterogeneous Adaptation for 3D Human Pose
• NeurIPS 20253DPE-Gaze:Unlocking the Potential of 3D Facial Priors for Generalized Gaze Estimation
• ICML 2025A General Representation-Based Approach to Multi-Source Domain Adaptation
• ICLR 2025Adaptive Energy Alignment for Accelerating Test-Time Adaptation
• CVPR 2025A4A: Adapter for Adapter Transfer via All-for-All Mapping for Cross-Architecture Models
• NeurIPS 2024A Cross-Domain Benchmark for Active Learning
• ICML 2024Batch Singular Value Polarization and Weighted Semantic Augmentation for Universal Domain Adaptation
• ICLR 2024Adapting Large Language Models via Reading Comprehension
• CVPR 2024Day-Night Cross-domain Vehicle Re-identification
• NeurIPS 2023(Almost) Provable Error Bounds Under Distribution Shift via Disagreement Discrepancy
• ICML 2023In Search for a Generalizable Method for Source Free Domain Adaptation
• ICLR 2023Addressing Parameter Choice Issues in Unsupervised Domain Adaptation by Aggregation
• CVPR 2023A New Benchmark: On the Utility of Synthetic Data With Blender for Bare Supervised Learning and Downstream Domain Adaptation
• NeurIPS 2026A unified pairwise distribution matching framework for graph domain adaptation under structure shift
• ICML 2026Content-Style Identification via Differential Independence
• ICLR 2026Beyond In-Domain Detection: SpikeScore for Cross-Domain Hallucination Detection
• CVPR 2026Bridge Basis-Driven Causal Inference Marries VFMs for Domain Generalization

全部域适应论文

共 81 篇，第 2 / 2 页

51. NeurIPS 2025PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation
52. NeurIPS 2025PANGEA: Projection-Based Augmentation with Non-Relevant General Data for Enhanced Domain Adaptation in LLMs
53. NeurIPS 2025Partition-Then-Adapt: Combating Prediction Bias for Reliable Multi-Modal Test-Time Adaptation
54. NeurIPS 2025Principled Model Routing for Unknown Mixtures of Source Domains
55. NeurIPS 2025ReservoirTTA: Prolonged Test-time Adaptation for Evolving and Recurring Domains
56. NeurIPS 2025Rethinking Entropy in Test-Time Adaptation: The Missing Piece from Energy Duality
57. NeurIPS 2025Rethinking Joint Maximum Mean Discrepancy for Visual Domain Adaptation
58. NeurIPS 2025RidgeLoRA: Matrix Ridge Enhanced Low-Rank Adaptation of Large Language Models
59. NeurIPS 2025RoME: Domain-Robust Mixture-of-Experts for MILP Solution Prediction across Domains
60. NeurIPS 2025RrED: Black-box Unsupervised Domain Adaptation via Rectifying-reasoning Errors of Diffusion
61. NeurIPS 2025S-Crescendo: A Nested Transformer Weaving Framework for Scalable Nonlinear System in S-Domain Representation
62. NeurIPS 2025Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation
63. NeurIPS 2025SGCD: Stain-Guided CycleDiffusion for Unsupervised Domain Adaptation of Histopathology Image Classification
64. NeurIPS 2025SNAP: Low-Latency Test-Time Adaptation with Sparse Updates
65. NeurIPS 2025Spatiotemporal Consensus with Scene Prior for Unsupervised Domain Adaptive Person Search
66. NeurIPS 2025Statistics Caching Test-Time Adaptation for Vision-Language Models
67. NeurIPS 2025StelLA: Subspace Learning in Low-rank Adaptation using Stiefel Manifold
68. NeurIPS 2025Synergy over Discrepancy: A Partition-Based Approach to Multi-Domain LLM Fine-Tuning
69. NeurIPS 2025Synthetic-powered predictive inference
70. NeurIPS 2025The Primacy of Magnitude in Low-Rank Adaptation
71. NeurIPS 2025Towards foundational LiDAR world models with efficient latent flow matching
72. NeurIPS 2025Towards Single-Source Domain Generalized Object Detection via Causal Visual Prompts
73. NeurIPS 2025Towards Unsupervised Domain Bridging via Image Degradation in Semantic Segmentation
74. NeurIPS 2025Towards Unsupervised Open-Set Graph Domain Adaptation via Dual Reprogramming
75. NeurIPS 2025Training-Free Test-Time Adaptation via Shape and Style Guidance for Vision-Language Models
76. NeurIPS 2025Transfer Faster, Price Smarter: Minimax Dynamic Pricing under Cross-Market Preference Shift
77. NeurIPS 2025Uncertainty-quantified Rollout Policy Adaptation for Unlabelled Cross-domain Video Temporal Grounding
78. NeurIPS 2025UniHG: A Large-scale Universal Heterogeneous Graph Dataset and Benchmark for Representation Learning and Cross-Domain Transferring
79. NeurIPS 2025Universal Cross-Tokenizer Distillation via Approximate Likelihood Matching
80. NeurIPS 2025Unleashing Foundation Vision Models: Adaptive Transfer for Diverse Data-Limited Scientific Domains
81. NeurIPS 2025Vicinity-Guided Discriminative Latent Diffusion for Privacy-Preserving Domain Adaptation

常见问题

2023–2026 年四大会议一共有多少篇域适应论文？

共 866 篇，按年份分别是 2023 年 174 篇、2024 年 209 篇、2025 年 220 篇、2026 年 263 篇。

NeurIPS 2026 有多少篇域适应论文？

NeurIPS 2026 有 78 篇域适应论文，上一届是 81 篇。

ICLR 2026 有多少篇域适应论文？

ICLR 2026 有 46 篇域适应论文，上一届是 45 篇。

ICML 2026 有多少篇域适应论文？

ICML 2026 有 69 篇域适应论文，上一届是 48 篇。

CVPR 2026 有多少篇域适应论文？

CVPR 2026 有 70 篇域适应论文，上一届是 46 篇。

哪个会议的域适应论文最多？

CVPR，2023–2026 年共 268 篇。