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 149 篇，第 1 / 3 页

1. ICLR 2025A Quantum Circuit-Based Compression Perspective for Parameter-Efficient Learning
2. ICLR 2025AdaRankGrad: Adaptive Gradient Rank and Moments for Memory-Efficient LLMs Training and Fine-Tuning
3. ICLR 2025Addax: Utilizing Zeroth-Order Gradients to Improve Memory Efficiency and Performance of SGD for Fine-Tuning Language Models
4. ICLR 2025ADePT: Adaptive Decomposed Prompt Tuning for Parameter-Efficient Fine-tuning
5. ICLR 2025Agent Skill Acquisition for Large Language Models via CycleQD
6. ICLR 2025Amulet: ReAlignment During Test Time for Personalized Preference Adaptation of LLMs
7. ICLR 2025An Auditing Test to Detect Behavioral Shift in Language Models
8. ICLR 2025Analyzing and Boosting the Power of Fine-Grained Visual Recognition for Multi-modal Large Language Models
9. ICLR 2025As Simple as Fine-tuning: LLM Alignment via Bidirectional Negative Feedback Loss
10. ICLR 2025Better autoregressive regression with LLMs via regression-aware fine-tuning
11. ICLR 2025Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation
12. ICLR 2025CBraMod: A Criss-Cross Brain Foundation Model for EEG Decoding
13. ICLR 2025ComLoRA: A Competitive Learning Approach for Enhancing LoRA
14. ICLR 2025Computational Limits of Low-Rank Adaptation (LoRA) Fine-Tuning for Transformer Models
15. ICLR 2025CONDA: Adaptive Concept Bottleneck for Foundation Models Under Distribution Shifts
16. ICLR 2025CryoFM: A Flow-based Foundation Model for Cryo-EM Densities
17. ICLR 2025CTSyn: A Foundation Model for Cross Tabular Data Generation
18. ICLR 2025DaWin: Training-free Dynamic Weight Interpolation for Robust Adaptation
19. ICLR 2025dEBORA: Efficient Bilevel Optimization-based low-Rank Adaptation
20. ICLR 2025Decoupling Angles and Strength in Low-rank Adaptation
21. ICLR 2025DELIFT: Data Efficient Language model Instruction Fine-Tuning
22. ICLR 2025Directional Gradient Projection for Robust Fine-Tuning of Foundation Models
23. ICLR 2025Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models
24. ICLR 2025Does Spatial Cognition Emerge in Frontier Models?
25. ICLR 2025DRESSing Up LLM: Efficient Stylized Question-Answering via Style Subspace Editing
26. ICLR 2025Dynamic Low-Rank Sparse Adaptation for Large Language Models
27. ICLR 2025DynaPrompt: Dynamic Test-Time Prompt Tuning
28. ICLR 2025Efficient Model Editing with Task-Localized Sparse Fine-tuning
29. ICLR 2025Efficiently Learning at Test-Time: Active Fine-Tuning of LLMs
30. ICLR 2025Emergence of meta-stable clustering in mean-field transformer models
31. ICLR 2025Emerging Safety Attack and Defense in Federated Instruction Tuning of Large Language Models
32. ICLR 2025Encryption-Friendly LLM Architecture
33. ICLR 2025Enhancing Cognition and Explainability of Multimodal Foundation Models with Self-Synthesized Data
34. ICLR 2025Enhancing Federated Domain Adaptation with Multi-Domain Prototype-Based Federated Fine-Tuning
35. ICLR 2025Enhancing Zeroth-order Fine-tuning for Language Models with Low-rank Structures
36. ICLR 2025Ensembles of Low-Rank Expert Adapters
37. ICLR 2025Exploring the Effectiveness of Object-Centric Representations in Visual Question Answering: Comparative Insights with Foundation Models
38. ICLR 2025Federated Residual Low-Rank Adaptation of Large Language Models
39. ICLR 2025Fine-Tuning Attention Modules Only: Enhancing Weight Disentanglement in Task Arithmetic
40. ICLR 2025Fine-tuning can Help Detect Pretraining Data from Large Language Models
41. ICLR 2025Fine-tuning with Reserved Majority for Noise Reduction
42. ICLR 2025Forget the Data and Fine-Tuning! Just Fold the Network to Compress
43. ICLR 2025Fundamental Limitations on Subquadratic Alternatives to Transformers
44. ICLR 2025Fundamental Limits of Prompt Tuning Transformers: Universality, Capacity and Efficiency
45. ICLR 2025Generative Adapter: Contextualizing Language Models in Parameters with A Single Forward Pass
46. ICLR 2025Generative Representational Instruction Tuning
47. ICLR 2025GeoLoRA: Geometric integration for parameter efficient fine-tuning
48. ICLR 2025GOAL: A Generalist Combinatorial Optimization Agent Learner
49. ICLR 2025HiRA: Parameter-Efficient Hadamard High-Rank Adaptation for Large Language Models
50. ICLR 2025Hyper-Connections

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。