# MoE in reinforcement learning and robotics

Area: [Mixture-of-experts](https://mlsanity.com/moe/index.md). HTML: https://mlsanity.com/moe/moe-in-reinforcement-learning-and-robotics/

## Key facts

- MoE in reinforcement learning and robotics: 151 papers on arXiv since 2018 (data through 14 Oct 2026).
- 9 of its 151 mapped papers (6%) cite “Mixtures of Experts Unlock Parameter Scaling for Deep RL” (2024, arXiv 2402.08609).
- 8 of its 151 mapped papers (5%) cite “DriveMoE: Mixture-of-Experts for Vision-Language-Action Model in End-to-End Autonomous Driving” (2025, arXiv 2505.16278).
- 4 new papers 17 Sep–14 Oct.

## Most cited here

- [Mixtures of Experts Unlock Parameter Scaling for Deep RL](https://mlsanity.com/paper/2402.08609/) (2024), cited by 9 papers here
- [DriveMoE: Mixture-of-Experts for Vision-Language-Action Model in End-to-End Autonomous Driving](https://mlsanity.com/paper/2505.16278/) (2025), cited by 8 papers here
- [MENTOR: Mixture-of-Experts Network with Task-Oriented Perturbation for Visual Reinforcement Learning](https://mlsanity.com/paper/2410.14972/) (2024), cited by 7 papers here
- [Group Sequence Policy Optimization](https://mlsanity.com/paper/2507.18071/) (2025), cited by 6 papers here
- [Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers](https://mlsanity.com/paper/2510.11370/) (2025), cited by 6 papers here
- [Mixture of Experts in a Mixture of RL settings](https://mlsanity.com/paper/2406.18420/) (2024), cited by 5 papers here
- [Probabilistic Mixture-of-Experts for Efficient Deep Reinforcement Learning](https://mlsanity.com/paper/2104.09122/) (2021), cited by 4 papers here
- [Kimi K2: Open Agentic Intelligence](https://mlsanity.com/paper/2507.20534/) (2025), cited by 4 papers here
- [MoE-Loco: Mixture of Experts for Multitask Locomotion](https://mlsanity.com/paper/2503.08564/) (2025), cited by 3 papers here
- [MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention](https://mlsanity.com/paper/2506.13585/) (2025), cited by 3 papers here

## New (17 Sep–14 Oct)

- [UniW(A)M: A Unified Mixture-of-Experts Framework for World and World-Action Models](https://mlsanity.com/paper/2610.18697/) (2026-10-13)
- [WARP-VLA: Wrist-Camera Adaptation for View-Robust Policy Execution in Vision-Language-Action Models](https://mlsanity.com/paper/2610.11508/) (2026-10-08)
- [TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models](https://mlsanity.com/paper/2610.07767/) (2026-10-06)
- [Structuring MoE Expert Selection for Agentic Reinforcement Learning](https://mlsanity.com/paper/2610.07332/) (2026-10-05)