RXed AI News

Your AI intelligence briefing
Education Saturday, May 23, 2026 7 posts
Microsoft's Fara1.5 agents (4B/9B/27B) score 72% on Online-Mind2Web, outperforming OpenAI and Gemini. @Microsoft https://www.marktechpost.com/2026/05/22/microsoft-releases-fara1-5-a-family-of-browser-computer-use-agents-4b-9b-27b-that-outperform-openai-operator-and-gemini-2-5-computer-use-on-online-mind2web/
· View on 𝕏
Deepseek’s founder tells investors AGI research comes before short-term profits amid $10B funding round. https://the-decoder.com/deepseek-reportedly-prioritizes-agi-research-over-quick-profits-despite-billions-in-funding/ Answer:
· View on 𝕏
NuExtract3 released: open-weight 4B VLM for Markdown, OCR, structured extraction, self-hostable https://www.reddit.com/r/MachineLearning/comments/1tkejqr/nuextract3_released_openweight_4b_vlm_for/
· View on 𝕏
LoREnc introduces Low-Rank Encryption to secure foundation models and LoRA adapters against IP leakage and model recovery attacks. https://huggingface.co/papers/2605.13163
· View on 𝕏
Garry Tan of @YCombinator launches GBrain, open-source memory layer for AI agents. https://www.marktechpost.com/2026/05/22/a-step-by-step-coding-tutorial-to-implement-gbrain-the-self-wiring-memory-layer-built-by-y-combinators-garry-tan-for-ai-agents/
· View on 𝕏
COLM 2026 reviews are live on Reddit—community feedback highlights mixed results and concerns about evaluation quality. https://www.reddit.com/r/MachineLearning/comments/1tkuu66/colm_2026_reviewsdiscussion_d/
· View on 𝕏
AutoRubric-T2I uses rule-based rewards to align text-to-image generation with human preferences. https://huggingface.co/papers/2605.17602
· View on 𝕏