Science ❯ Computer Science ❯ Artificial Intelligence ❯ Vision-Language Models

Multimodal Reasoning

Microsoft Releases Phi‑4‑Reasoning‑Vision‑15B on Foundry and Hugging Face

The 15B‑parameter model adds prompt‑controlled multi‑step reasoning to high‑resolution visual perception.

RAG Research Moves to Graph, Lightweight, and Agentic Designs as New Benchmarks Steer Fixes