“What is the best LLM for RAG and grounded long-form inputs? @GoogleDeepMind DeepMind FACTS is a new benchmark for how well LLMs can generate factually accurate, long-form responses while remaining faithful to provided source documents. 👀 TL;DR; 📊 1,719 examples (860 public,
https://x.com/_philschmid/status/1869052954579189976
“Notion, the much-loved workspace platform, is redefining productivity with Notion AI and Cohere Rerank. Every search and Notion AI interaction now runs through Rerank for fast and relevant results. Rerank delivers: – Enhanced search accuracy – Unmatched cost efficiency – A
https://x.com/cohere/status/1868666666411786696
“I’ll get straight to the point. We trained 2 new models. Like BERT, but modern. ModernBERT. Not some hypey GenAI thing, but a proper workhorse model, for retrieval, classification, etc. Real practical stuff. It’s much faster, more accurate, longer context, and more useful. 🧵
https://x.com/jeremyphoward/status/1869786023963832509?s=46
“…and Slack Enterprise Search supercharges Agentforce with: 🔍 Permission-aware search across DMs, channels, and files 📊 Insights drawn from connected apps 🤖 Retrieval-Augmented Generation (RAG) for smarter responses. Finding signal from noise is one of the most challenging
https://x.com/MatthewBerman/status/1869128243863040240
