Multi-Turn Interaction RL and Credit Assignment (Split)

New Locations

This page remains as an entry point for old links. Its original content has been split into two sections by topic:

• Formalization (multi-turn MDPs, POMDPs, trajectory structure, and action masks) → 19.2 Multi-Turn Reinforcement Learning
• Credit assignment (ORM/PRM/SALT, turn-level discounting, and the Mini Agent Loop experiment) → 19.3 Trajectory Credit Assignment

Use the two links above to open the corresponding material.