Mage-VL: An Efficient Codec-Native Streaming Multimodal Foundation Model
Standard vision-language models (VLMs) suffer from Moravec's paradox: they excel at complex offline visual reasoning but struggle with simple streaming perception tasks and process them inefficiently. We present Mage-VL, an efficient codec-native streaming foundation model for real-time multimodal understanding and interaction. At its core, our custom tokenizer, Mage-ViT, replaces uniform frame sampling by selectively encoding dynamic, entropy-rich regions using motion vectors and residual energy across sparse anchor (I) and predicted (P) frames. Operating at a 16 x 16 patch level, this reduces visual token consumption by over 75% while preserving spatiotemporal context. Trained from scra...
| Authors | Senqiao Yang, Kaichen Zhang, Zhaoyang Jia et al. |
|---|---|
| Primary category | cs.CV |
| Published | 27 July 2026 |
| Last updated at source | 27 July 2026 |
Popularity, last 2 days
0 now. Counts as reported by the source, snapshotted daily by Silytics.
Connected in the knowledge graph
- microsoft/Mage-VLModel · cited by
Models and datasets referencing this paper
- microsoft/Mage-VLModel
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.