Research16d ago
Stanford & UC Berkeley Introduce Librarian Agents: Training LLMs to Proactively Curate Workspaces, Cutting Navigation Costs by 33.8%
Researchers from Stanford University and UC Berkeley introduced Librarian Agents (arXiv:2610.22273). To overcome the bottleneck of reactive RAG methods struggling with complex cross-document reasoning at query time, the framework uses reinforcement learning to train LLMs to perform anticipatory directed curation on messy long-context workspaces into structured synthetic filesystems. On BrowseComp+, it hits 55.86% accuracy while reducing navigation token consumption by 33.8% to 51%.
Tooling33d ago
Cloudflare AI Search goes GA: native multimodal embeddings + PDF OCR; billing from Nov 1; hybrid search default
On 2026-10-01 Cloudflare made AI Search generally available: managed index/retrieval (Workers AI + Vectorize + R2 + Browser Run) ships native multimodal embeddings (Qwen3-VL-Embedding), OCR for scanned PDFs, 10 MiB text/OCR PDF limits, hybrid search by default, and billing from 2026-11-01 with a free allotment.
Models28d ago
Perplexity open-sources pplx-embed-v2-late: 0.6B/9B multimodal ColBERT retrievers, OCR-free PDFs, small model queries the big model's index
On Oct 7 Perplexity released pplx-embed-v2-late, multimodal late-interaction (ColBERT) retrievers in 0.6B (340M active) and 9B (7.4B active) sizes, built on Qwen3.5 with bidirectional attention. Each token gets a 128-dim vector scored by MaxSim, covering text, images and rendered PDFs or slides without OCR. Both sizes share one embedding space, so the 0.6B model can query a 9B-built index. Public ViDoRe v3 image nDCG@10 is 62.3% / 65.2%; MIT-licensed weights are on Hugging Face.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.