sample-efficiency

StraTA proposes strategic trajectory abstraction for long-horizon LLM agents, using hierarchical GRPO-style rollout with diverse strategy sampling and critical self-judgment to improve sample efficiency and final performance over frontier models and prior RL baselines.

FreshPER introduces a freshness-aware prioritized experience replay method for LLM/VLM reinforcement learning that addresses the 'priority staleness' problem by applying exponential age decay to stored priorities, enabling off-policy reuse of trajectories. Evaluated on eight agentic, reasoning, and math tasks, FreshPER significantly outperforms on-policy baselines with gains up to +367% on Sokoban.

OpenAI co-organizes the MineRL 2020 Competition to advance sample-efficient reinforcement learning algorithms that leverage human demonstrations. Participants compete to obtain a diamond in Minecraft using only 8 million simulator samples and 4 days of single-GPU training, with access to a 60+ million frame human demonstration dataset.

OpenAI presents a method for training AI agents using human preference feedback, where an agent learns reward functions from human comparisons of behavior trajectories and uses reinforcement learning to optimize for the inferred goals. The approach demonstrates strong sample efficiency, requiring less than 1000 bits of human feedback to train an agent to perform a backflip.