SOURCE-LINKED INTELLIGENCE

On the Design Fundamentals of Pixel Text Representation Learning

arXiv · AI, language, vision and robotics · article · Sep 1, 2026 · UTC

Text-rich visual inputs require models that can read, retrieve, and compress language directly in pixel space, yet existing pixel-text encoders struggle with fixed resolution pretraining, visual shortcut learning, weak visual grounding, and multilingual visual text understanding. In this work, we investigate the fundamental design principles required for robust visual text representation learning. Through systematic controlled ablations, we identify four critical components: variable image resolutions and rendered font sizes provide spatial proxies for high-resolution document generalization;

Read original source ↗ Open in workspace

recordType

paper

region

Global