RADIO1D: Elastic Representations for Condensed Vision Modeling
This paper challenges the assumption that vision-language models (VLMs) require fixed patch-based 2D vision features. Analyzing fine-tuned vision encoders, we find that representations become increasingly abstract and less spatially coherent during VLM training. Notably, models trained with image-text alignment (such as SigLIP2) develop a small number of specialized tokens that effectively summarize global image content. Building on this, we introduce RADIO1D, which compresses images into a compact, variable-length 1D token sequence using multi-teacher knowledge distillation and an autoencoder design. The resulting representations exhibit strong hierarchical summarization, enabling accura...
| Authors | Greg Heinrich, Mike Ranzinger, Collin McCarthy et al. |
|---|---|
| Primary category | cs.CV |
| Published | 3 July 2026 |
| Last updated at source | 3 July 2026 |
Connected in the knowledge graph
- nvidia/C-RADIOv4-1D-HModel · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.