SOURCE-LINKED INTELLIGENCE

Studying Image Tokenizers as Visual Languages in Unified Multimodal Models

arXiv · AI, language, vision and robotics · article · Sep 8, 2026 · UTC

Image tokenizers define the ``visual language'' of unified multimodal models, yet are commonly studied through isolated metrics or generation-/understanding-only evaluations. These evaluations do not fully capture how visual tokens behave when modeled jointly with text. We build a controlled pure-autoregressive testbed and track task-specific validation losses during multimodal continual pretraining across text, image, text-to-image (T2I), and image-to-text (I2T) prediction. We examine how these losses scale and relate to downstream performance, then use them to study multimodal learnability--

Read original source ↗ Open in workspace

recordType

paper

region

Global