Must-read papers of the week: 13 on robot foundation models, looped models, agent memory
TheTuringPost · x · 2026-10-13
TheTuringPost's weekly paper roundup spans robotics (RoboJEPA, ARC reasoning recipe for robot foundation models, Long-WAM, RealtimeWAM), agents and memory (Memento 3 recursive self-improvement, REMORY residual memory for context compaction, AgentGarten, RoboQuest), and architecture/optimization (LoGRA low-rank RL scaling, looped models at fixed points, self-retrospection distillation, LLM-discovered molecular relaxation algorithms).
More from Research
- Deployable AI Workshop at AAAI 2027 opens submissions, deadline Nov 20, 2026 — ravi_iitm · 2026-10-13
- Study: AI-generated books flood Amazon self-publishing and crowd out human authors' sales — TuhinChakr · 2026-10-13
- MosaiChunk: compositing spatio-temporal KV memory for autoregressive video generation — camenduru · 2026-10-13
- Jev-like Decision Models Spark a Wave of Papers: Papers with Code Roundup — zainhas · 2026-10-13
- RLVR produces "evolved modularity" AIs; 3D reasoning gains confirmed as RL-trained, not emergent — jessi_cata · 2026-10-13
- Is your model swap actually worse, or just noise? Open-source Assay brings stats to LLM evals — Professional-Run3614 · 2026-10-13