Must-read papers of the week: Agent Lightning, LEGO-RL and 14 more picks
TheTuringPost · x · 2026-08-25
Turing Post's weekly must-read AI papers, spanning agentic RL, memory, and long-horizon tasks:
- Agentic RL training: Agent Lightning v1.0, LEGO-RL (harness-native RL for coding agents), EnvHarness (awakening static worlds), SPADE (self-play in adaptive synthetic executable environments)
- Long-horizon skills: SkillGate (in-policy skill selection), Looped LMs for compositional tool calling, Graph Engineering for LLM agents
- Memory & continual learning: Chain-of-Experience, cross-model memory transfer via target-side reader adaptation
- Others: CLEAR (utility-preserving safety routing), EXIMO (VLM-guided VLA exploration), FreeToken (edge-native MoE serving), Llama-Mobile (2.7-bit VLM quantization), matrix multiplication exponent improvements
More from coding & agent
- Building a Custom Obsidian Search with Claude Code in Five Minutes — evielync · 2026-08-25
- FactoryAI workflow: Transcribe chats to generate designs instantly — matanSF · 2026-08-25
- Voice Agent Testing Methodology: Separating ASR Failures from Downstream Logic — Admirable-Wallaby457 · 2026-08-25
- Ace launches as an agentic teammate with its own Mac mini, email, and iPhone — zan2434 · 2026-08-25
- Building a layer to freeze intent and verify authority inheritance for long-lived agents — tallmetommy · 2026-08-25
- The AI flywheel: from task acceleration to an AI chief of staff — leebase65 · 2026-08-25