SERO augments long-horizon agentic RL with executable oracle continuations
Shanghai-AI-Laboratory · hf · 2026-10-16
Shanghai AI Laboratory, Shanghai Jiao Tong University, and Tongji University introduce SERO (State-Rooted Executable Rollout Optimization) for long-horizon agentic reinforcement learning.
- Problem: long-horizon agentic RL is constrained by the futures visited by the current policy — when rollouts fail to reach useful behaviors, better credit assignment can't supervise states and actions never encountered
- On-policy distillation (OPD) and multi-turn variants only schedule, replay, refine, or validate teacher guidance; they don't use the oracle-controlled future itself as the training trajectory
- SERO restores a state actually reached by the current policy and augments training with executable oracle continuations, weighted by policy compatibility, with no inference-time overhead
More from coding & agent
- Tencent eval agents hit Amap 2,818 times hunting zoo entrances, digging for leaked API keys — Upper_Collection3335 · 2026-10-16
- Fintech HR built a sourcing agent on 60,000 past applicants — 30 qualified candidates in a month — G0NZ0BOY · 2026-10-16
- Orchestrating a 100-image portrait of humanity for aliens with ChatGPT, Argon and p5.brush — dx8xb · 2026-10-16
- 3x DGX Spark cluster runs GLM-5.3-Flash locally behind a proxy, 1.75M tokens/day at 119W — Teknium · 2026-10-16
- Builder's one rule for trusted desktop agents: read freely, but sending, deleting, buying always ask first — No_Fan5310 · 2026-10-16
- GPT-built unifi-simulator makes MikroTik switches appear inside UniFi Network — Pytlicek · 2026-10-16