From Trainee to Trainer: LLM-Designed Training Environment for RL with Multi-Agent Reasoning

• URL: http://arxiv.org/abs/2606.17682v1
• Date: Tue, 16 Jun 2026 08:48:58 GMT
• Title: From Trainee to Trainer: LLM-Designed Training Environment for RL with Multi-Agent Reasoning
• Abstract summary: Reinforcement learning pipelines for Large Language Model (LLM) training often rely on manually redesigned environments between stages.<n>We propose the LLM-as-Environment-Engineer framework in which the current policy model analyzes failure trajectories together with contextual information.<n>We also introduce MAPF-FrozenLake, a controllable testbed whose generator exposes multi-dimensional environment configurations.
• Score: 24.54315594953994
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning pipelines for Large Language Model (LLM) training often rely on manually redesigned environments between stages, requiring practitioners to heuristically infer which configuration will best improve the current policy. To automate this process, we propose the LLM-as-Environment-Engineer framework in which the current policy model analyzes failure trajectories together with contextual information and proposes modifications to the next-stage training environment configuration. We also introduce MAPF-FrozenLake, a controllable testbed whose generator exposes multi-dimensional environment configurations, making it suitable for studying and benchmarking environment redesign. On this testbed, we condition the environment engineer on structured summaries of policy behavior, failure cases, and environment statistics, from which it produces the configuration for the next training stage. With Qwen3-4B as the backbone, our framework achieves the strongest aggregate performance on our benchmarks, outperforming larger proprietary LLMs (e.g., GPT, Gemini) and fixed-environment training baselines. We further analyze which forms of context are most effective, finding that successful environment updates rely on failure evidence and preserve configurations that already work. Interestingly, the current RL checkpoint serves as a better environment engineer than the original base model, suggesting that policy learning improves the model's ability to diagnose its remaining weaknesses.

Related papers

• Path Planning through Multi-Agent Reinforcement Learning in Dynamic Environments [2.116865312302264]
  We propose a scalable, region-aware reinforcement learning framework for path planning in dynamic environments.<n>Our method builds on the observation that environmental changes, although dynamic, are often localized within bounded regions.
  arXiv  Detail & Related papers  (2025-11-19T09:48:44Z)
• Simulating Environments with Reasoning Models for Agent Training [55.98861707136674]
  Building bespoke environments for training is heavy, brittle, and limits progress.<n>We propose two frameworks: Simia-SFT and Simia-RL.<n>Simia-SFT and Simia-RL enable scalable agent training without environment engineering.
  arXiv  Detail & Related papers  (2025-11-03T18:29:57Z)
• Sample-Efficient Online Learning in LM Agents via Hindsight Trajectory Rewriting [92.57796055887995]
  We introduce ECHO, a prompting framework that adapts hindsight experience replay from reinforcement learning for language model agents.<n> ECHO generates optimized trajectories for alternative goals that could have been achieved during failed attempts.<n>We evaluate ECHO on stateful versions of XMiniGrid, a text-based navigation and planning benchmark, and PeopleJoinQA, a collaborative information-gathering enterprise simulation.
  arXiv  Detail & Related papers  (2025-10-11T18:11:09Z)
• Feedback-Driven Tool-Use Improvements in Large Language Models via Automated Build Environments [70.42705564227548]
  We propose an automated environment construction pipeline for large language models (LLMs)<n>This enables the creation of high-quality training environments that provide detailed and measurable feedback without relying on external tools.<n>We also introduce a verifiable reward mechanism that evaluates both the precision of tool use and the completeness of task execution.
  arXiv  Detail & Related papers  (2025-08-12T09:45:19Z)
• MLE-Dojo: Interactive Environments for Empowering LLM Agents in Machine Learning Engineering [57.156093929365255]
  Gym-style framework for systematically reinforcement learning, evaluating, and improving autonomous large language model (LLM) agents.<n>MLE-Dojo covers diverse, open-ended MLE tasks carefully curated to reflect realistic engineering scenarios.<n>Its fully executable environment supports comprehensive agent training via both supervised fine-tuning and reinforcement learning.
  arXiv  Detail & Related papers  (2025-05-12T17:35:43Z)
• Training Agents with Weakly Supervised Feedback from Large Language Models [19.216542820742607]
  This paper introduces a novel training method for LLM-based agents using weakly supervised signals from a critic LLM.<n>Our agents are trained in iterative manner, where they initially generate trajectories through environmental interaction.<n>Tests on the API-bank dataset show consistent improvement in our agents' capabilities and comparable performance to GPT-4.
  arXiv  Detail & Related papers  (2024-11-29T08:47:04Z)
• Enhancing the Hierarchical Environment Design via Generative Trajectory Modeling [8.256433006393243]
  We introduce a hierarchical MDP framework for environment design under resource constraints. It consists of an upper-level RL teacher agent that generates suitable training environments for a lower-level student agent. Our proposed method significantly reduces the resource-intensive interactions between agents and environments.
  arXiv  Detail & Related papers  (2023-09-30T08:21:32Z)
• Online Reinforcement Learning in Non-Stationary Context-Driven Environments [12.954992692713898]
  We study online reinforcement learning (RL) in non-stationary environments.<n>Online RL is challenging in such environments due to "catastrophic forgetting" (CF)<n>We present Locally Constrained Policy Optimization (LCPO), an online RL approach that combats CF by anchoring policy outputs on old experiences.
  arXiv  Detail & Related papers  (2023-02-04T15:31:19Z)
• Learning to Continuously Optimize Wireless Resource in a Dynamic Environment: A Bilevel Optimization Perspective [52.497514255040514]
  This work develops a new approach that enables data-driven methods to continuously learn and optimize resource allocation strategies in a dynamic environment. We propose to build the notion of continual learning into wireless system design, so that the learning model can incrementally adapt to the new episodes. Our design is based on a novel bilevel optimization formulation which ensures certain fairness" across different data samples.
  arXiv  Detail & Related papers  (2021-05-03T07:23:39Z)
• Learning to Continuously Optimize Wireless Resource In Episodically Dynamic Environment [55.91291559442884]
  This work develops a methodology that enables data-driven methods to continuously learn and optimize in a dynamic environment. We propose to build the notion of continual learning into the modeling process of learning wireless systems. Our design is based on a novel min-max formulation which ensures certain fairness" across different data samples.
  arXiv  Detail & Related papers  (2020-11-16T08:24:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.