Long-horizon LLM agents interact with their environments through complete assistant turns while generating each action token by token. We observe that critic values vary primarily across turns, motiva
Multiscale Policy Optimization improves long-horizon agent training by aligning credit assignment and policy updates at the turn scale.
Long-horizon LLM agents interact with their environments through complete assistant turns while generating each action token by token. We observe that critic values vary primarily across turns, motivating optimization anchored at turns with finer credit allocation among tokens. We propose Multiscale Policy Optimization (MPO), combining Multiscale Advantage Estimation (MAE) and turn-level importance sampling to align credit assignment and policy updates at the turn scale. Extensive experiments across search, tool-use, and automation tasks with 4B and 30B-A3B models demonstrate consistent gains over strong RL baselines.
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
Multiscale Policy Optimization for Long-Horizon LLM Agents was first seen on Hugging Face · Daily papers on Oct 28, 2026. Since then it has appeared in 21 checks on 3 different days. Its best position was #1 of 30, reached on Oct 28, 23:28 UTC. It dropped off the list after Oct 30, 05:28 UTC.
Position on Hugging Face · Daily papers at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hugging Face.
Archived daily, weekly and monthly lists that include this entry.
Not at the moment. It was last on the list on Oct 30, 05:28 UTC, after 3 days of trending.
#1 on Hugging Face · Daily papers, on Oct 28, 23:28 UTC. Trendoday checks this list every 120 minutes.
It was first seen on Oct 28, 2026 and has been on the list on 3 days (21 checks) since.