Cover Image

Impact of Pretraining Word Co-occurrence on Compositional Generalization in Multimodal Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Qu, Helen, Xie, Sang Michael
Format:	Preprint
Published:	2025
Subjects:	Computer Vision and Pattern Recognition Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Connect Later: Improving Fine-tuning for Robustness with Targeted Augmentations
by: Qu, Helen, et al.
Published: (2024)

Pretrain Where? Investigating How Pretraining Data Diversity Impacts Geospatial Foundation Model Performance
by: Kaur, Amandeep, et al.
Published: (2026)

LMFusion: Adapting Pretrained Language Models for Multimodal Generation
by: Shi, Weijia, et al.
Published: (2024)

Context-Aware Multimodal Pretraining
by: Roth, Karsten, et al.
Published: (2024)

Towards Knowledge Guided Pretraining Approaches for Multimodal Foundation Models: Applications in Remote Sensing
by: Ravirathinam, Praveen, et al.
Published: (2024)

Foundation Model-oriented Robustness: Robust Image Model Evaluation with Pretrained Models
by: Zhang, Peiyan, et al.
Published: (2023)

A Practitioner's Guide to Continual Multimodal Pretraining
by: Roth, Karsten, et al.
Published: (2024)

Draw Like an Artist: Complex Scene Generation with Diffusion Model via Composition, Painting, and Retouching
by: Liu, Minghao, et al.
Published: (2024)

Pretrained Visual Uncertainties
by: Kirchhof, Michael, et al.
Published: (2024)

SpectralEarth-FM: Bringing Hyperspectral Imagery into Multimodal Earth Observation Pretraining
by: Braham, Nassim Ait Ali, et al.
Published: (2026)

Simplifying Knowledge Transfer in Pretrained Models
by: Jain, Siddharth, et al.
Published: (2025)

COMPRER: A Multimodal Multi-Objective Pretraining Framework for Enhanced Medical Image Representation
by: Lutsker, Guy, et al.
Published: (2024)

Is Large-Scale Pretraining the Secret to Good Domain Generalization?
by: Teterwak, Piotr, et al.
Published: (2024)

Why Fine-grained Labels in Pretraining Benefit Generalization?
by: Hong, Guan Zhe, et al.
Published: (2024)

SILMM: Self-Improving Large Multimodal Models for Compositional Text-to-Image Generation
by: Qu, Leigang, et al.
Published: (2024)

No "Zero-Shot" Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance
by: Udandarao, Vishaal, et al.
Published: (2024)

Pretrained Diffusion Models Are Inherently Skipped-Step Samplers
by: Xu, Wenju
Published: (2025)

Pretrained Image-Text Models are Secretly Video Captioners
by: Zhang, Chunhui, et al.
Published: (2025)

Progressive Compositionality in Text-to-Image Generative Models
by: Han, Evans Xu, et al.
Published: (2024)

Fantastic Gains and Where to Find Them: On the Existence and Prospect of General Knowledge Transfer between Any Pretrained Model
by: Roth, Karsten, et al.
Published: (2023)

2.5 Years in Class: A Multimodal Textbook for Vision-Language Pretraining
by: Zhang, Wenqi, et al.
Published: (2025)

Reflecting on the State of Rehearsal-free Continual Learning with Pretrained Models
by: Thede, Lukas, et al.
Published: (2024)

ROSE: Revolutionizing Open-Set Dense Segmentation with Patch-Wise Perceptual Large Multimodal Model
by: Han, Kunyang, et al.
Published: (2024)

Negative Label Guided OOD Detection with Pretrained Vision-Language Models
by: Jiang, Xue, et al.
Published: (2024)

Image Clustering via the Principle of Rate Reduction in the Age of Pretrained Models
by: Chu, Tianzhe, et al.
Published: (2023)

Accelerating Augmentation Invariance Pretraining
by: Lin, Jinhong, et al.
Published: (2024)

What Variables Affect Out-of-Distribution Generalization in Pretrained Models?
by: Harun, Md Yousuf, et al.
Published: (2024)

Learning Multimodal Latent Generative Models with Energy-Based Prior
by: Yuan, Shiyu, et al.
Published: (2024)

Enhancing Compositional Generalization via Compositional Feature Alignment
by: Wang, Haoxiang, et al.
Published: (2024)

Words That Make Language Models Perceive
by: Wang, Sophie L., et al.
Published: (2025)

Breaking the Likelihood-Quality Trade-off in Diffusion Models by Merging Pretrained Experts
by: Esfandiari, Yasin, et al.
Published: (2025)

Compositional Generalization Requires Linear, Orthogonal Representations in Vision Embedding Models
by: Uselis, Arnas, et al.
Published: (2026)

Medical Semantic Segmentation with Diffusion Pretrain
by: Li, David, et al.
Published: (2025)

Polyp Segmentation Generalisability of Pretrained Backbones
by: Sanderson, Edward, et al.
Published: (2024)

Multimodal Generalized Category Discovery
by: Su, Yuchang, et al.
Published: (2024)

Rethinking Generative Image Pretraining: How Far Are We From Scaling Up Next-Pixel Prediction?
by: Yan, Xinchen, et al.
Published: (2025)

Contrastive Test-Time Composition of Multiple LoRA Models for Image Generation
by: Meral, Tuna Han Salih, et al.
Published: (2024)

STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation
by: Shen, Ying, et al.
Published: (2026)

Self-Supervised Pretraining for Aerial Road Extraction
by: Polley, Rupert, et al.
Published: (2025)

Object-level Self-Distillation for Vision Pretraining
by: Hızlı, Çağlar, et al.
Published: (2025)