Vision-Language Pretraining papers, 2023–2026

Key findings

• Vision-Language Pretraining went from 65 papers in 2023 to 87 in 2026, roughly 1.3x in three years.
• NeurIPS publishes the most Vision-Language Pretraining work: 93 papers from 2023 to 2026, 35% of the four-venue total.
• The fastest-growing overlap is with Applications · computer vision: papers tagged with both rose from 0 in 2023 to 9 in 2026.
• Among the 5 topics under Open-Vocabulary Vision, it ranks #4 by 2026 paper count; the fastest grower in the group, Visual Grounding, is up 4.3x versus 1.3x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026Benchmarking Optimizers for Large Language Model Pretraining
• ICML 2026How Does the Pretraining Distribution Shape In-Context Learning? A Fundamental Trade-Off
• ICLR 2026PonderLM: Pretraining Language Models to Ponder in Continuous Space
• CVPR 2026Large-scale Codec Avatars The Unreasonable Effectiveness of Large-scale Avatar Pretraining
• NeurIPS 2025ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretraining
• ICML 2025Soup-of-Experts: Pretraining Specialist Models via Parameters Averaging
• ICLR 2025Attributing Culture-Conditioned Generations to Pretraining Corpora
• CVPR 2025MAP: Unleashing Hybrid Mamba-Transformer Vision Backbone's Potential with Masked Autoregressive Pretraining
• NeurIPS 2024Data curation via joint example selection further accelerates multimodal learning
• ICML 2024Position: Considerations for Differentially Private Learning with Large-Scale Public Pretraining
• ICLR 2024Demystifying CLIP Data
• CVPR 2024ParameterNet: Parameters Are All You Need for Large-scale Visual Pretraining of Mobile Networks
• NeurIPS 2023Bootstrapping Vision-Language Learning with Decoupled Language Pre-training
• ICML 2023Contrastive Learning Meets Homophily: Two Birds with One Stone
• ICLR 2023Energy-Inspired Self-Supervised Pretraining for Vision Models
• CVPR 2023Active Finetuning: Exploiting Annotation Budget in the Pretraining-Finetuning Paradigm
• NeurIPS 2026C3P: Contrastive promoter-protein pretraining yields representations capturing bacterial gene regulation
• ICML 2026Multipole Semantic Attention: A Fast Approximation of Softmax Attention for Pretraining
• ICLR 2026MICLIP: Learning to Interpret Representation in Vision Models
• CVPR 2026The Surprising Effectiveness of Noise Pretraining for Implicit Neural Representations

All Vision-Language Pretraining papers

263 papers, page 3 of 6

101. ICLR 2025Beyond Random Augmentations: Pretraining with Hard Views
102. ICLR 2025Compositional Entailment Learning for Hyperbolic Vision-Language Models
103. ICLR 2025Detecting Backdoor Samples in Contrastive Language Image Pretraining
104. ICLR 2025Efficient stagewise pretraining via progressive subnetworks
105. ICLR 2025Fine-tuning can Help Detect Pretraining Data from Large Language Models
106. ICLR 2025Generalization v.s. Memorization: Tracing Language Models’ Capabilities Back to Pretraining Data
107. ICLR 2025Harnessing Diversity for Important Data Selection in Pretraining Large Language Models
108. ICLR 2025Holographic Node Representations: Pre-training Task-Agnostic Node Embeddings
109. ICLR 2025Improving Pretraining Data Using Perplexity Correlations
110. ICLR 2025Latent Action Pretraining from Videos
111. ICLR 2025Probabilistic Language-Image Pre-Training
112. ICLR 2025Self-Introspective Decoding: Alleviating Hallucinations for Large Vision-Language Models
113. ICLR 2025TIPS: Text-Image Pretraining with Spatial awareness
114. ICLR 2025VTDexManip: A Dataset and Benchmark for Visual-tactile Pretraining and Dexterous Manipulation with Reinforcement Learning
115. ICML 2025A Bayesian Model Selection Criterion for Selecting Pretraining Checkpoints
116. ICML 2025Learning Invariant Causal Mechanism from Vision-Language Models
117. ICML 2025Soup-of-Experts: Pretraining Specialist Models via Parameters Averaging
118. NeurIPS 20254D-VLA: Spatiotemporal Vision-Language-Action Pretraining with Cross-Scene Calibration
119. NeurIPS 2025ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretraining
120. NeurIPS 2025AmorLIP: Efficient Language-Image Pretraining via Amortization
121. NeurIPS 2025Analyzing Similarity Metrics for Data Selection for Language Model Pretraining
122. NeurIPS 2025Born a Transformer -- Always a Transformer? On the Effect of Pretraining on Architectural Abilities
123. NeurIPS 2025CLIPTTA: Robust Contrastive Vision-Language Test-Time Adaptation
124. NeurIPS 2025Contrastive Self-Supervised Learning As Neural Manifold Packing
125. NeurIPS 2025Critical Batch Size Revisited: A Simple Empirical Approach to Large-Batch Language Model Training
126. NeurIPS 2025EgoDTM: Towards 3D-Aware Egocentric Video-Language Pretraining
127. NeurIPS 2025Enhancing Contrastive Learning with Variable Similarity
128. NeurIPS 2025Exploring Polyglot Harmony: On Multilingual Data Allocation for Large Language Models Pretraining
129. NeurIPS 2025Filter Like You Test: Data-Driven Data Filtering for CLIP Pretraining
130. NeurIPS 2025From Faults to Features: Pretraining to Learn Robust Representations against Sensor Failures
131. NeurIPS 2025From Pretraining to Pathology: How Noise Leads to Catastrophic Inheritance in Medical Models
132. NeurIPS 2025Group-Level Data Selection for Efficient Pretraining
133. NeurIPS 2025Meta CLIP 2: A Worldwide Scaling Recipe
134. NeurIPS 2025MuRating: A High Quality Data Selecting Approach to Multilingual Large Language Model Pretraining
135. NeurIPS 2025Mysteries of the Deep: Role of Intermediate Representations in Out of Distribution Detection
136. NeurIPS 2025Object Concepts Emerge from Motion
137. NeurIPS 2025Object-centric binding in Contrastive Language-Image Pretraining
138. NeurIPS 2025Perception Encoder: The best visual embeddings are not at the output of the network
139. NeurIPS 2025REVE: A Foundation Model for EEG - Adapting to Any Setup with Large-Scale Pretraining on 25,000 Subjects
140. NeurIPS 2025The Mirage of Performance Gains: Why Contrastive Decoding Fails to Mitigate Object Hallucinations in MLLMs?
141. NeurIPS 2025Tracing the Representation Geometry of Language Models from Pretraining to Post-training
142. NeurIPS 2025UniDomain: Pretraining a Unified PDDL Domain from Real-World Demonstrations for Generalizable Robot Task Planning
143. NeurIPS 2025VITRIX-UniViTAR: Unified Vision Transformer with Native Resolution
144. CVPR 2024Anchor-based Robust Finetuning of Vision-Language Models
145. CVPR 2024BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning
146. CVPR 2024Building Vision-Language Models on Solid Foundations with Masked Distillation
147. CVPR 2024Efficient Vision-Language Pre-training by Cluster Masking
148. CVPR 2024EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything
149. CVPR 2024FFF: Fixing Flawed Foundations in Contrastive Pre-Training Results in Very Strong Vision-Language Models
150. CVPR 2024Iterated Learning Improves Compositionality in Large Vision-Language Models

FAQ

How many Vision-Language Pretraining papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

263 in total: 65 in 2023, 55 in 2024, 56 in 2025, 87 in 2026.

How many Vision-Language Pretraining papers are at NeurIPS 2026?

NeurIPS 2026 has 27 Vision-Language Pretraining papers, versus 26 the year before.

How many Vision-Language Pretraining papers are at ICLR 2026?

ICLR 2026 has 14 Vision-Language Pretraining papers, versus 17 the year before.

How many Vision-Language Pretraining papers are at ICML 2026?

ICML 2026 has 19 Vision-Language Pretraining papers, versus 3 the year before.

How many Vision-Language Pretraining papers are at CVPR 2026?

CVPR 2026 has 27 Vision-Language Pretraining papers, versus 10 the year before.

Which conference publishes the most Vision-Language Pretraining papers?

NeurIPS, with 93 papers from 2023 to 2026.