视觉语言预训练论文趋势 2023–2026

对比预训练

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 263 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

视觉语言预训练论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
2023241382065
2024161291855
2025261731056
20262714192787

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

全部视觉语言预训练论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 263 篇，第 5 / 6 页

  1. CVPR 2023CLIP2: Contrastive Language-Image-Point Pretraining From Real-World Point Cloud Data
  2. CVPR 2023CREPE: Can Vision-Language Foundation Models Reason Compositionally?
  3. CVPR 2023DegAE: A New Pretraining Paradigm for Low-Level Vision
  4. CVPR 2023Finetune Like You Pretrain: Improved Finetuning of Zero-Shot Vision Models
  5. CVPR 2023FlowFormer++: Masked Cost Volume Autoencoding for Pretraining Optical Flow Estimation
  6. CVPR 2023HyperMatch: Noise-Tolerant Semi-Supervised Learning via Relaxed Contrastive Constraint
  7. CVPR 2023Image as a Foreign Language: BEiT Pretraining for Vision and Vision-Language Tasks
  8. CVPR 2023Is BERT Blind? Exploring the Effect of Vision-and-Language Pretraining on Visual Language Understanding
  9. CVPR 2023Learning Imbalanced Data With Vision Transformers
  10. CVPR 2023Learning Instance-Level Representation for Large-Scale Multi-Modal Pretraining in E-Commerce
  11. CVPR 2023MaskCLIP: Masked Self-Distillation Advances Contrastive Language-Image Pretraining
  12. CVPR 2023MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers
  13. CVPR 2023OmniMAE: Single Model Masked Pretraining on Images and Videos
  14. CVPR 2023Open-Set Fine-Grained Retrieval via Prompting Vision-Language Evaluator
  15. CVPR 2023Q: How To Specialize Large Vision-Language Models to Data-Scarce VQA Tasks? A: Self-Train on Unlabeled Images!
  16. CVPR 2023Region-Aware Pretraining for Open-Vocabulary Object Detection With Vision Transformers
  17. CVPR 2023Training Debiased Subnetworks With Contrastive Weight Pruning
  18. CVPR 2023VILA: Learning Image Aesthetics From User Comments With Vision-Language Pretraining
  19. ICLR 2023A theoretical study of inductive biases in contrastive learning
  20. ICLR 2023CodeBPE: Investigating Subtokenization Options for Large Language Model Pretraining on Source Code
  21. ICLR 2023Contrastive Alignment of Vision to Language Through Parameter-Efficient Transfer Learning
  22. ICLR 2023Energy-Inspired Self-Supervised Pretraining for Vision Models
  23. ICLR 2023From $t$-SNE to UMAP with contrastive learning
  24. ICLR 2023HiCLIP: Contrastive Language-Image Pretraining with Hierarchy-aware Attention
  25. ICLR 2023Learning Language Representations with Logical Inductive Bias
  26. ICLR 2023Proposal-Contrastive Pretraining for Object Detection from Fewer Data
  27. ICLR 2023Protein Representation Learning by Geometric Structure Pretraining
  28. ICLR 2023Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints
  29. ICLR 2023UL2: Unifying Language Learning Paradigms
  30. ICLR 2023What Do Self-Supervised Vision Transformers Learn?
  31. ICLR 2023When and Why Vision-Language Models Behave like Bags-Of-Words, and What to Do About It?
  32. ICML 2023Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
  33. ICML 2023Contrastive Learning Meets Homophily: Two Birds with One Stone
  34. ICML 2023End-to-end Training of Deep Boltzmann Machines by Unbiased Contrastive Divergence with Local Mode Initialization
  35. ICML 2023Lifelong Language Pretraining with Distribution-Specialized Experts
  36. ICML 2023Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding
  37. ICML 2023Pretraining Language Models with Human Preferences
  38. ICML 2023Why Is Public Pretraining Necessary for Private Model Training?
  39. ICML 2023XTab: Cross-table Pretraining for Tabular Transformers
  40. NeurIPS 2023Bootstrapping Vision-Language Learning with Decoupled Language Pre-training
  41. NeurIPS 2023Brain encoding models based on multimodal transformers can transfer across language and vision
  42. NeurIPS 2023CoDet: Co-occurrence Guided Region-Word Alignment for Open-Vocabulary Object Detection
  43. NeurIPS 2023Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP
  44. NeurIPS 2023D4: Improving LLM Pretraining via Document De-Duplication and Diversification
  45. NeurIPS 2023DiffVL: Scaling Up Soft Body Manipulation using Vision-Language Driven Differentiable Physics
  46. NeurIPS 2023Don’t Stop Pretraining? Make Prompt-based Fine-tuning Powerful Learner
  47. NeurIPS 2023DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining
  48. NeurIPS 2023Enhancing Adversarial Contrastive Learning via Adversarial Invariant Regularization
  49. NeurIPS 2023Frequency-Enhanced Data Augmentation for Vision-and-Language Navigation
  50. NeurIPS 2023Image Captioners Are Scalable Vision Learners Too

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查视觉语言预训练在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "vision-language-pretraining"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

常见问题

2023–2026 年四大会议一共有多少篇视觉语言预训练论文？
共 263 篇，按年份分别是 2023 年 65 篇、2024 年 55 篇、2025 年 56 篇、2026 年 87 篇。
NeurIPS 2026 有多少篇视觉语言预训练论文？
NeurIPS 2026 有 27 篇视觉语言预训练论文，上一届是 26 篇。
ICLR 2026 有多少篇视觉语言预训练论文？
ICLR 2026 有 14 篇视觉语言预训练论文，上一届是 17 篇。
ICML 2026 有多少篇视觉语言预训练论文？
ICML 2026 有 19 篇视觉语言预训练论文，上一届是 3 篇。
CVPR 2026 有多少篇视觉语言预训练论文？
CVPR 2026 有 27 篇视觉语言预训练论文，上一届是 10 篇。
哪个会议的视觉语言预训练论文最多？
NeurIPS，2023–2026 年共 93 篇。

「开放词汇视觉」下的其他主题