Turing Post's Must-Read Papers of the Week: Cortex, Librarian Agents, Mosaic and 12 More
TheTuringPost · x · 2026-10-20
Turing Post rounds up the week's must-read AI papers spanning model memory, agents and open-ended research evaluation:
- Cortex: consolidating context into parametric memory (ByteDance Seed et al.; on a 14B model it recovers 88.9% of context-induced gains after 50 sequential factual edits while retaining 99.8% of non-target performance)
- Librarian Agents, Persistent Planar Memory, Execution Before Memory: new work on agent memory and execution
- OpenGCUA and HexaShell: thin harnesses exploring VLM agents as robot coding agents
- Mosaic: multi-session agentic reinforcement learning
- Into the Unknown: a new benchmark testing AI agents on open scientific research problems
- Plus DuoDiscover, WAVE-0, ACT3D, relational learning as world modeling and more
The full list with links and a weekly AI news digest is included.
More from coding & agent
- REA gains 214K stars in a week, tops GitHub's fastest-growing AI repos — sven_ai · 2026-10-20
- Dev goes 'graph pilled': graphs will anchor every company's world model — _AustinCalvert_ · 2026-10-20
- RuneBench: Benchmarking AI Coding Agents by Playing RuneScape — charliermarsh · 2026-10-20
- New Claude Cookbook: Auto-Review Agent Session Transcripts for Security Issues — brada · 2026-10-20
- Conversational Video Editing: Driving CapCut Web with Codex for Marketing Videos — Div_pradeep · 2026-10-20
- Claude Opus reverse-engineers Revolut's full design system from one URL — CurieuxExplorer · 2026-10-20