MPO: turn-scale credit assignment for long-horizon LLM agents
IQuestLab · hf · 2026-10-28
Researchers from Peking University, Tsinghua, IQuest Research, Renmin University and HKU propose Multiscale Policy Optimization (MPO) for long-horizon LLM agents. Key observation: critic values vary mainly at the turn scale. MPO combines Multiscale Advantage Estimation (MAE) with turn-level importance sampling, anchoring credit assignment and policy updates at the turn scale with finer token-level allocation, validated across search, tool-use and automation tasks with 4B+ models.
More from coding & agent
- Claude Catches North Korean Supply Chain Attack Disguised as an Auth Error — banteg · 2026-10-28
- Expo ships agent skills collection for AI coding tools, already 3.3k GitHub stars — expo · 2026-10-28
- Does a Google Play uploader MCP exist, and would you trust an agent to ship to production? — Extender7777 · 2026-10-28
- Lucas Meijer: writing and evaling skills forces you to say what you care about — lucasmeijer · 2026-10-28
- Merge CEO Shensi Ding stays hands-on with Codex: execs set the AI fluency ceiling — OpenAI · 2026-10-28
- Detection Isn't Blocking: How to Actually Verify Your AI Agent's Security Guard — shoumikgoswami · 2026-10-28