context-aware reward

criteria generation (1) dynamics learning (1) task inference (1) large language model (1) task uncertainty (1) task adaptation (1) dynamics model (1) policy adaptation (1) meta-reinforcement learning (1) reward hacking (1)