Megadose Built for builders and researchers.

Feed Uncut

Home ResearchAverage-Reward Policy Identification from a Single Continuin

Average-Reward Policy Identification from a Single Continuing Trajectory

By Tongxi Wang

· ArXiv · AI/CL/LG · Oct 14, 2026

ArXiv · AI/CL/LG's note

score 4

Categories: Research

Megadose home · About · Sources · Privacy · Contact · Sitemap

Megadose tracks AI progress across model releases, benchmarks, OSS, research, and product launches. Updated every 30 minutes.