news

Mar 12, 2026 Entropy-Preserving Reinforcement Learning appears at ICLR 2026.
Feb 05, 2025 Robust Autonomy Emerges from Self-Play (GIGAFLOW): robust driving from 1.6 billion km of self-play, without human data. Published at ICML 2025.
Feb 03, 2025 Reinforcement Learning for Long-Horizon Interactive LLM Agents (LOOP) is out on arXiv.
May 23, 2023 Proximal Policy Gradient Arborescence for quality-diversity RL is out on arXiv; spotlight at ICLR 2024.
May 20, 2023 DexPBT: scaling up dexterous manipulation for hand-arm systems with population-based training, at RSS 2023.