# FARE: Fast-Slow Agentic Robotic Exploration

arXiv: https://arxiv.org/abs/2601.14681 · HTML: https://mlsanity.com/paper/2601.14681/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use.
- Builds on 18 of its 30 references that are in the map.
- First posted on arXiv 21 Jan 2026.

## Placed in

- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)

## Builds on (most cited first, 18 of 18)

- [Graph of Thoughts: Solving Elaborate Problems with Large Language Models](https://mlsanity.com/paper/2308.09687/) (2023)
- [Exploring the Potential of Large Language Models (LLMs) in Learning on Graphs](https://mlsanity.com/paper/2307.03393/) (2023)
- [GPT4Graph: Can Large Language Models Understand Graph Structured Data ? An Empirical Evaluation and Benchmarking](https://mlsanity.com/paper/2305.15066/) (2023)
- [FAST-LIO2: Fast Direct LiDAR-inertial Odometry](https://mlsanity.com/paper/2107.06829/) (2021)
- [LLaGA: Large Language and Graph Assistant](https://mlsanity.com/paper/2402.08170/) (2024)
- [GraphLLM: Boosting Graph Reasoning Ability of Large Language Model](https://mlsanity.com/paper/2310.05845/) (2023)
- [InstructGraph: Boosting Large Language Models via Graph-centric Instruction Tuning and Preference Alignment](https://mlsanity.com/paper/2402.08785/) (2024)
- [GITA: Graph to Visual and Textual Integration for Vision-Language Graph Reasoning](https://mlsanity.com/paper/2402.02130/) (2024)
- [Autonomous Exploration Under Uncertainty via Deep Reinforcement Learning on Graphs](https://mlsanity.com/paper/2007.12640/) (2020)
- [ARiADNE: A Reinforcement learning approach using Attention-based Deep Networks for Exploration](https://mlsanity.com/paper/2301.11575/) (2023)
- [Deep Reinforcement Learning-based Large-scale Robot Exploration](https://mlsanity.com/paper/2403.10833/) (2024)
- [Autonomous UAV Exploration of Dynamic Environments via Incremental Sampling and Probabilistic Roadmap](https://mlsanity.com/paper/2010.07429/) (2020)
- [Explore-Bench: Data Sets, Metrics and Evaluations for Frontier-based and Deep-reinforcement-learning-based Autonomous Exploration](https://mlsanity.com/paper/2202.11931/) (2022)
- [FIG-OP: Exploring Large-Scale Unknown Environments on a Fixed Time Budget](https://mlsanity.com/paper/2203.06316/) (2022)
- [MAexp: A Generic Platform for RL-based Multi-Agent Exploration](https://mlsanity.com/paper/2404.12824/) (2024)
- [Fast Think-on-Graph: Wider, Deeper and Faster Reasoning of Large Language Model on Knowledge Graph](https://mlsanity.com/paper/2501.14300/) (2025)
- [HPHS: Hierarchical Planning based on Hybrid Frontier Sampling for Unknown Environments Exploration](https://mlsanity.com/paper/2407.10660/) (2024)
- [HEADER: Hierarchical Robot Exploration via Attention-Based Deep Reinforcement Learning with Expert-Guided Reward](https://mlsanity.com/paper/2510.15679/) (2025)

## Built on by (most cited first, 1 of 1)

- [Driver-WM: A Driver-Centric Traffic-Conditioned Latent World Model for In-Cabin Dynamics Rollout](https://mlsanity.com/paper/2605.05092/) (2026)

## Abstract (arXiv)

This work advances autonomous robot exploration by integrating agent-level semantic reasoning with fast local control. We introduce FARE, a hierarchical autonomous exploration framework that integrates a large language model (LLM) for global reasoning with a reinforcement learning (RL) policy for local decision making. FARE follows a fast-slow thinking paradigm. The slow-thinking LLM module interprets a concise textual description of the unknown environment and synthesizes an agent-level exploration strategy, which is then grounded into a sequence of global waypoints through a topological graph. To further improve reasoning efficiency, this module employs a modularity-based pruning mechanism that reduces redundant graph structures. The fast-thinking RL module executes exploration by reacting to local observations while being guided by the LLM-generated global waypoints. The RL policy is additionally shaped by a reward term that encourages adherence to the global waypoints, enabling coherent and robust closed-loop behavior. This architecture decouples semantic reasoning from geometric decision, allowing each module to operate in its appropriate temporal and spatial scale. In challenging simulated environments, our results show that FARE achieves substantial improvements in exploration efficiency over state-of-the-art baselines. We further deploy FARE on hardware and validate it in complex, large scale 200m×130m building environment.