# Phase-Aware Mixture of Experts for Agentic Reinforcement Learning

arXiv: https://arxiv.org/abs/2602.17038 · HTML: https://mlsanity.com/paper/2602.17038/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 32 of its 37 references that are in the map.
- First posted on arXiv 19 Feb 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)
- Mixture-of-experts / [MoE in reinforcement learning and robotics](https://mlsanity.com/moe/moe-in-reinforcement-learning-and-robotics/)

## Builds on (most cited first, 30 of 32)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [High-Dimensional Continuous Control Using Generalized Advantage Estimation](https://mlsanity.com/paper/1506.02438/) (2015)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding](https://mlsanity.com/paper/2006.16668/) (2020)
- [Gradient Surgery for Multi-Task Learning](https://mlsanity.com/paper/2001.06782/) (2020)
- [WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents](https://mlsanity.com/paper/2207.01206/) (2022)
- [ALFWorld: Aligning Text and Embodied Environments for Interactive Learning](https://mlsanity.com/paper/2010.03768/) (2020)
- [GLaM: Efficient Scaling of Language Models with Mixture-of-Experts](https://mlsanity.com/paper/2112.06905/) (2021)
- [GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks](https://mlsanity.com/paper/1711.02257/) (2017)
- [Group-in-Group Policy Optimization for LLM Agent Training](https://mlsanity.com/paper/2505.10978/) (2025)
- [Conflict-Averse Gradient Descent for Multi-task Learning](https://mlsanity.com/paper/2110.14048/) (2021)
- [The Option-Critic Architecture](https://mlsanity.com/paper/1609.05140/) (2016)
- [ST-MoE: Designing Stable and Transferable Sparse Expert Models](https://mlsanity.com/paper/2202.08906/) (2022)
- [AgentTuning: Enabling Generalized Agent Abilities for LLMs](https://mlsanity.com/paper/2310.12823/) (2023)
- [FireAct: Toward Language Agent Fine-tuning](https://mlsanity.com/paper/2310.05915/) (2023)
- [Recurrent Independent Mechanisms](https://mlsanity.com/paper/1909.10893/) (2019)
- [Modular Multitask Reinforcement Learning with Policy Sketches](https://mlsanity.com/paper/1611.01796/) (2016)
- [Learning Modular Neural Network Policies for Multi-Task and Multi-Robot Transfer](https://mlsanity.com/paper/1609.07088/) (2016)
- [MATEval: A Multi-Agent Discussion Framework for Advancing Open-Ended Text Evaluation](https://mlsanity.com/paper/2403.19305/) (2024)
- [MoniTor: Exploiting Large Language Models with Instruction for Online Video Anomaly Detection](https://mlsanity.com/paper/2510.21449/) (2025)

## Built on by (most cited first, 6 of 6)

- [MoRSE: Task-Oriented Multi-Agent System with Mixture of Role-Subtask Experts](https://mlsanity.com/paper/2608.09251/) (2026)
- [SCA: Spatial Credit Assignment for Reinforcement Learning of GUI Agents](https://mlsanity.com/paper/2609.36939/) (2026)
- [Reinforcement Learning Enhanced LLM Agents for Complex Vehicle Routing Problems](https://mlsanity.com/paper/2609.00859/) (2026)
- [SRPO: Setwise Relative Policy Optimization for Multi-Agent LLMs](https://mlsanity.com/paper/2609.08452/) (2026)
- [Choosing Before Acting: Comparative Value Estimation for Long-Horizon Tool-Use Agents](https://mlsanity.com/paper/2610.02330/) (2026)
- [Credit Where It Matters: Dependency-Aware Policy Optimization for Terminal Agents](https://mlsanity.com/paper/2610.03634/) (2026)

## Abstract (arXiv)

Reinforcement learning (RL) has equipped LLM agents with a strong ability to solve complex tasks. However, existing RL methods normally use a single policy network, causing simplicity bias where simple tasks occupy most parameters and dominate gradient updates, leaving insufficient capacity for complex tasks. A plausible remedy could be employing the Mixture-of-Experts (MoE) architecture in the policy network, as MoE allows different parameters (experts) to specialize in different tasks, preventing simple tasks from dominating all parameters. However, a key limitation of traditional MoE is its token-level routing, where the router assigns each token to specialized experts, which fragments phase-consistent patterns into scattered expert assignments and thus undermines expert specialization. In this paper, we propose Phase-Aware Mixture of Experts (PA-MoE). It first features a lightweight phase router that learns latent phase boundaries directly from the RL objective without pre-defining phase categories. Then, the phase router allocates temporally consistent assignments to the same expert, allowing experts to preserve phase-specific expertise. Experimental results demonstrate the effectiveness of our proposed PA-MoE.