HomeFlow: A Data Flywheel for Smart Home Agent Training with Verifiable Simulation

• URL: http://arxiv.org/abs/2606.01230v1
• Date: Sun, 31 May 2026 13:26:46 GMT
• Title: HomeFlow: A Data Flywheel for Smart Home Agent Training with Verifiable Simulation
• Abstract summary: We propose HomeFlow to generate high-quality training data for smart home agents.<n>HomeFlow uses HomeEnv as a unified simulation environment and HomeMaker to procedurally generate diverse home settings.<n>HomeFlow-RL-4B and HomeFlow-RL-8B achieve task success rates of 84.60% and 87.03%, respectively.
• Score: 22.87553318476621
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language model agents are moving beyond text-only interaction toward physical-world control, with smart homes as a representative domain. Real domestic interaction requires understanding ambiguous intents, operating in dynamic environments, and performing multi-turn reasoning. However, existing methods struggle to generate high-quality training data for smart home agents. We propose HomeFlow, a verifiable data flywheel for this domain. HomeFlow uses HomeEnv as a unified simulation environment and HomeMaker to procedurally generate diverse home settings. Subsequently, Blueprint compiles open-ended user intents into executable state-based success conditions, while MCTS-Flow synthesizes diverse, verifiable multi-turn trajectories through environment-guided tree search. We then optimize the agents via supervised fine-tuning and step-wise RLVE, which facilitates iterative improvement through authentic physical feedback. We further construct SmartHome-Bench to evaluate the agent across various smart home tasks. On this benchmark, HomeFlow-RL-4B and HomeFlow-RL-8B achieve task success rates of 84.60% and 87.03%. It is worth noting that HomeFlow-RL-8B even surpasses the leading GPT-5.5 by 1.23 percentage points.

Related papers

• SPADE: Self-Play in Adaptive Synthetic Executable Environments [110.80471334001574]
  We introduce SPADE, a self-play RL framework in which a single LLM plays two roles.<n>The Environment Designer writes complete, long-horizon training environments as executable code with an OpenAI Gym-style reset()/step() interface.<n>We find several components critical to success: grounding the Environment Designer on documents sampled from a large pretraining corpus, and giving it an accumulated environment memory.
  arXiv  Detail & Related papers  (2026-08-19T17:58:56Z)
• OdysSim: Building Foundation Models for Human Behavior Simulation [70.35265860287608]
  Large language models are increasingly deployed as human simulators for interactive evaluation and social simulation.<n>Yet helpfulness-driven post-training pulls them toward a homogeneous, overly agreeable assistant register.<n>We present OdysSim, the largest open systematic investigation of behavioral foundation models.
  arXiv  Detail & Related papers  (2026-06-12T07:31:55Z)
• IR-SIM: A Lightweight Skill-Native Simulator for Navigation, Learning, and Benchmarking [59.6657632820611]
  IR-SIM is a skill-native navigation simulator designed for rapid scenario construction, benchmarking, and robot learning.<n> scenarios are entirely defined by YAML configuration files that specify mobile robot kinematics, geometric collision checking, LiDAR sensing, visualization, and behavior modules.<n>The experiments showcase the convenience and versatility of IR-SIM in multiple tasks.
  arXiv  Detail & Related papers  (2026-06-07T16:55:37Z)
• Omni-SimpleMem: Autoresearch-Guided Discovery of Lifelong Multimodal Agent Memory [76.63021613850093]
  We deploy an autonomous research pipeline to discover Omni-SimpleMem, a unified multimodal memory framework for lifelong AI agents.<n>System achieves state-of-the-art on two benchmarks, improving F1 by +411% on LoCoMo and +214% on Mem-Gallery.<n>We provide a taxonomy of six discovery types and identify four properties that make multimodal memory particularly suited for autoresearch.
  arXiv  Detail & Related papers  (2026-04-01T15:06:23Z)
• 6GAgentGym: Tool Use, Data Synthesis, and Agentic Learning for Network Management [4.299007466612815]
  6G-Forge bootstraps closed-loop training trajectories from NS-3 seeds via iterative Self-Instruct generation with execution verification against the Experiment Model.<n>Supervised fine-tuning on the resulting corpus followed by reinforcement learning with online closed-loop interaction enables an 8B open-source model to achieve comparable overall success rate to GPT-5.
  arXiv  Detail & Related papers  (2026-03-31T12:21:00Z)
• S5-HES Agent: Society 5.0-driven Agentic Framework to Democratize Smart Home Environment Simulation [1.4680035572775534]
  This paper presents the Society 5.0-driven Smart Home Environment Simulator Agent (S5-HES Agent)<n>The framework coordinates specialized agents through interchangeable large language models (LLMs)<n>A retrieval-augmented generation (RAG) pipeline with semantic, keyword, and hybrid search retrieves smart home knowledge.
  arXiv  Detail & Related papers  (2026-03-02T07:30:09Z)
• Dyna-Mind: Learning to Simulate from Experience for Better AI Agents [62.21219817256246]
  We argue that current AI agents need ''vicarious trial and error'' - the capacity to mentally simulate alternative futures before acting.<n>We introduce Dyna-Mind, a two-stage training framework that explicitly teaches (V)LM agents to integrate such simulation into their reasoning.
  arXiv  Detail & Related papers  (2025-10-10T17:30:18Z)
• SimuHome: A Temporal- and Environment-Aware Benchmark for Smart Home LLM Agents [10.925787133867951]
  SimuHome is a time-accelerated home environment that simulates smart devices, supports API calls, and reflects changes in environmental variables.<n>SimuHome provides a high-fidelity environment, and agents validated in SimuHome can be deployed on real Matter-compliant devices with minimal adaptation.<n>Our evaluation of 11 agents under a unified ReAct framework reveals that while models perform well on simple tasks, they struggle with latent intent inference, state verification, and especially temporal scheduling.
  arXiv  Detail & Related papers  (2025-09-29T04:54:20Z)
• AgentSense: Virtual Sensor Data Generation Using LLM Agents in Simulated Home Environments [2.9698967396161624]
  We leverage the idea of embodied AI agents-virtual agents that perceive and act within simulated environments guided by internal world models.<n>We introduce AgentSense, a virtual data generation pipeline in which agents live out daily routines in simulated smart homes.<n>Our approach produces rich, privacy-preserving sensor data that reflects real-world diversity.
  arXiv  Detail & Related papers  (2025-06-13T13:31:08Z)
• Waymax: An Accelerated, Data-Driven Simulator for Large-Scale Autonomous Driving Research [76.93956925360638]
  Waymax is a new data-driven simulator for autonomous driving in multi-agent scenes. It runs entirely on hardware accelerators such as TPUs/GPUs and supports in-graph simulation for training. We benchmark a suite of popular imitation and reinforcement learning algorithms with ablation studies on different design decisions.
  arXiv  Detail & Related papers  (2023-10-12T20:49:15Z)
• User Behavior Simulation with Large Language Model based Agents [116.74368915420065]
  We propose an LLM-based agent framework and design a sandbox environment to simulate real user behaviors. Based on extensive experiments, we find that the simulated behaviors of our method are very close to the ones of real humans.
  arXiv  Detail & Related papers  (2023-06-05T02:58:35Z)
• Parallel Reinforcement Learning Simulation for Visual Quadrotor Navigation [4.597465975849579]
  Reinforcement learning (RL) is an agent-based approach for teaching robots to navigate within the physical world. We present a simulation framework, built on AirSim, which provides efficient parallel training. Building on this framework, Ape-X is modified to incorporate decentralised training of AirSim environments.
  arXiv  Detail & Related papers  (2022-09-22T15:27:42Z)
• Towards Optimal Strategies for Training Self-Driving Perception Models in Simulation [98.51313127382937]
  We focus on the use of labels in the synthetic domain alone. Our approach introduces both a way to learn neural-invariant representations and a theoretically inspired view on how to sample the data from the simulator. We showcase our approach on the bird's-eye-view vehicle segmentation task with multi-sensor data.
  arXiv  Detail & Related papers  (2021-11-15T18:37:43Z)
• Sim-to-Real Transfer for Vision-and-Language Navigation [70.86250473583354]
  We study the problem of releasing a robot in a previously unseen environment, and having it follow unconstrained natural language navigation instructions. Recent work on the task of Vision-and-Language Navigation (VLN) has achieved significant progress in simulation. To assess the implications of this work for robotics, we transfer a VLN agent trained in simulation to a physical robot.
  arXiv  Detail & Related papers  (2020-11-07T16:49:04Z)

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.