基础模型适配论文趋势 2023–2026

主要结论

• 基础模型适配论文从 2023 年的 159 篇增长到 2026 年的 602 篇，三年间扩大到约 3.8 倍。
• NeurIPS 是发基础模型适配论文最多的会议，2023–2026 年共 531 篇，占四个会议总量的 36%。
• 交叉方向里涨得最快的是深度学习 · foundation models：同时属于两个主题的论文从 2023 年的 7 篇增至 2026 年的 66 篇。
• 在「元学习与迁移」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的持续学习是 1.7 倍，基础模型适配是 3.8 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 2026(Mis)generalization of Helpful-Only Fine-Tuning
• ICML 2026Active Learning with Foundation Model Priors: Efficient Learning under Class Imbalance
• ICLR 2026Benchmarking ECG Foundational Models: A Reality Check Across Clinical Tasks
• CVPR 2026CD-Buffer Complementary Dual-Buffer Framework for Test-Time Adaptation in Adverse Weather
• NeurIPS 2025An Investigation of Memorization Risk in Healthcare Foundation Models
• ICML 2025A Bayesian Model Selection Criterion for Selecting Pretraining Checkpoints
• ICLR 2025Better autoregressive regression with LLMs via regression-aware fine-tuning
• CVPR 20255%>100%: Breaking Performance Shackles of Full Fine-Tuning on Visual Recognition Tasks
• NeurIPS 2024Instruction Tuning Large Language Models to Understand Electronic Health Records
• ICML 2024By Tying Embeddings You Are Assuming the Distributional Hypothesis
• ICLR 2024A Hard-to-Beat Baseline for Training-free CLIP-based Adaptation
• CVPR 2024DePT: Decoupled Prompt Tuning
• NeurIPS 2023A Fast and Accurate Estimator for Large Scale Linear Model via Data Averaging
• ICML 2023Are Neurons Actually Collapsed? On the Fine-Grained Structure in Neural Representations
• ICLR 2023A Closer Look at Model Adaptation using Feature Distortion and Simplicity Bias
• CVPR 2023A-La-Carte Prompt Tuning (APT): Combining Distinct Data via Composable Prompting
• NeurIPS 2026A Graph Foundation Model for Unified Clustering
• ICML 2026AdaMeZO: Adam-style Zeroth-Order Optimizer for LLM Fine-tuning Without Maintaining the Moments
• ICLR 2026Bridging Radiology and Pathology Foundation Models via Concept-Based Multimodal Co-Adaptation
• CVPR 2026Fed-ADE Adaptive Learning Rate for Federated Post-adaptation under Distribution Shift

全部基础模型适配论文

共 91 篇，第 2 / 2 页

51. CVPR 2026Interactive Tracking A Human-in-the-Loop Paradigm with Memory-Augmented Adaptation
52. CVPR 2026IP-Adapter Is All You Need Towards Fine-Tuning-Free Diffusion-Based Talking Face
53. CVPR 2026Learning to Refuse Refusal-Aware Reinforcement Fine-Tuning for Hard-Irrelevant Queries in
54. CVPR 2026LEMON A Large Endoscopic MONocular Dataset and Foundation Model for
55. CVPR 2026LiteVGGT Boosting Vanilla VGGT via Geometry-aware Cached Token Merging
56. CVPR 2026Local Precise Refinement A Dual-Gated Mixture-of-Experts for Enhancing Foundation Model
57. CVPR 2026LoFA Learning to Predict Personalized Prior for Fast Adaptation of
58. CVPR 2026LoPrune Efficient Data Pruning for LoRA-Based Fine-Tuning of Vision Transformer
59. CVPR 2026Lyapunov Probes for Hallucination Detection in Large Foundation Models
60. CVPR 2026Mining Attribute Subspaces for Efficient Fine-tuning of 3D Foundation Models
61. CVPR 2026MOFA-VTON More Fashion Possibilities with Fine-Grained Adaptations in Virtual Try-On
62. CVPR 2026Multimodal Protein Language Models for Enzyme Kinetic Parameters From Substrate
63. CVPR 2026NitroGen An Open Foundation Model for Generalist Gaming Agents
64. CVPR 2026OccAny Generalized Unconstrained Urban 3D Occupancy
65. CVPR 2026ORION ORthonormal Text Encoding for Universal VLM AdaptatION
66. CVPR 2026ORSATR-X A Foundation Model based on Differential-and-Excitation Networks for Optical
67. CVPR 2026Parameter-Efficient Adaptation for MLLMs via Implicit Modality Decomposition
68. CVPR 2026PointTPA Dynamic Network Parameter Adaptation for 3D Scene Understanding
69. CVPR 2026Probabilistic Prompt Adaptation for Unified Image Aesthetics and Quality Assessment
70. CVPR 2026Rank-Guided Pseudo-Bias Learning for Robust Black-Box Adaptation
71. CVPR 2026ReFTA Breaking the Weight Reconstruction Bottleneck in Tensorized Parameter-Efficient Fine-Tuning
72. CVPR 2026Representation-Steered Incremental Adapter-Tuning for Class-Incremental Learning with Pre-Trained Models
73. CVPR 2026Rethinking BCE Loss for Multi-Label Image Recognition with Fine-Tuning
74. CVPR 2026Revisiting 2D Foundation Models for Scalable 3D Medical Image Classification
75. CVPR 2026Revisiting Model Stitching In the Foundation Model Era
76. CVPR 2026S2FT Parameter-Efficient Fine-Tuning in Sparse Spectrum Domain
77. CVPR 2026Scaling Dense Event-Stream Pretraining from Visual Foundation Models
78. CVPR 2026Selective Regularized and Calibrated Harnessing Vision Foundation Models for Cross-Domain
79. CVPR 2026SHAPE Structure-aware Hierarchical Unsupervised Domain Adaptation with Plausibility Evaluation for
80. CVPR 2026SoccerMaster A Vision Foundation Model for Soccer Understanding
81. CVPR 2026SOTA Self-adaptive Optimal Transport for Zero-Shot Classification with Multiple Foundation
82. CVPR 2026Taming Noise-Induced Prototype Degradation for Privacy-Preserving Personalized Federated Fine-Tuning
83. CVPR 2026TANGO Learning Distribution-wise Foundation Prior Consistency and Instance-wise Style Calibration
84. CVPR 2026Tell2Adapt A Unified Framework for Source Free Unsupervised Domain Adaptation
85. CVPR 2026The Blind Spot of Adaptation Quantifying and Mitigating Forgetting in
86. CVPR 2026Towards Realistic and Consistent Orbital Video Generation via 3D Foundation
87. CVPR 2026UniSER A Foundation Model for Unified Soft Effects Removal
88. CVPR 2026Universal 3D Shape Matching via Coarse-to-Fine Language Guidance
89. CVPR 2026Vision Foundation Models Can Be Good Tokenizers for Latent Diffusion
90. CVPR 2026Weight Space Representation Learning via Neural Field Adaptation
91. CVPR 2026WiTTA-Bench Benchmarking Test-Time Adaptation for WiFi Sensing

常见问题

2023–2026 年四大会议一共有多少篇基础模型适配论文？

共 1,492 篇，按年份分别是 2023 年 159 篇、2024 年 270 篇、2025 年 461 篇、2026 年 602 篇。

NeurIPS 2026 有多少篇基础模型适配论文？

NeurIPS 2026 有 182 篇基础模型适配论文，上一届是 174 篇。

ICLR 2026 有多少篇基础模型适配论文？

ICLR 2026 有 187 篇基础模型适配论文，上一届是 149 篇。

ICML 2026 有多少篇基础模型适配论文？

ICML 2026 有 142 篇基础模型适配论文，上一届是 76 篇。

CVPR 2026 有多少篇基础模型适配论文？

CVPR 2026 有 91 篇基础模型适配论文，上一届是 62 篇。

哪个会议的基础模型适配论文最多？

NeurIPS，2023–2026 年共 531 篇。