AdaHome: An Adaptive Smart Home Assistant using Local Small Language Models

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Eu Jin Lim, Zhaoxing Li, Sebastian Stein
• Venue: arXiv: cs.AI
• Published: 2026-07-20
• Collected: 2026-07-21
• URL: http://arxiv.org/abs/2607.18034v1
• PDF: https://arxiv.org/pdf/2607.18034v1
• DOI: UNKNOWN

Plain-Language Brief

A paper titled 'AdaHome: An Adaptive Smart Home Assistant using Local Small Language Models' presents a new smart home assistant system designed to work with locally deployed small language models in resource-constrained environments. The system, named AdaHome, improves upon existing approaches by introducing an intent-aware planning framework and a preference adaptation mechanism for personalization.

What the Document Says

• The paper presents AdaHome, an adaptive smart home assistant.
• AdaHome is designed for locally deployed small language models in smart home environments.
• It introduces an intent-aware planning framework.
• For commands requiring interpretation, it uses a Chain-of-Draft strategy.
• It proposes a preference adaptation mechanism for personalization.
• The authors evaluate AdaHome against LLM-based baselines under a unified small model setting.

Key Concepts Mentioned

• AdaHome
• smart home assistant
• local small language models
• intent-aware planning framework
• Chain-of-Draft strategy
• preference adaptation mechanism
• evaluation

Claims Made by the Document

• AdaHome achieves higher accuracy on direct commands (86.7%) and reduces latency by up to 3 times.
• It maintains competitive performance on ambiguous inputs with lower computational cost.
• In multi-turn scenarios, AdaHome achieves 88% preference consistency.

Evidence Used

• AdaHome achieves substantially higher accuracy on direct commands (86.7%) while reducing latency by up to 3$ imes$.
• Furthermore, it maintains competitive performance on ambiguous inputs with lower computational cost.
• In multi-turn scenarios, AdaHome achieves 88% preference consistency, compared to 52.5% for a prompt augmentation baseline.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Smart home assistants interpret a wide range of user commands, from explicit device control to underspecified and preference dependent requests. While recent systems based on Large Language Models (LLMs) improve this capability, they often rely on heavyweight reasoning pipelines and cloud-based deployment, limiting their efficiency and suitability for resource-constrained environments, and raising privacy concerns. In addition, existing approaches provide limited support for stable long-term personalization. To address these issues, we present AdaHome, an adaptive smart home assistant designed for locally deployed small language models in smart home environments. Rather than applying complex reasoning uniformly, AdaHome introduces an intent-aware planning framework that dynamically routes commands either to straightforward prompt-based or lightweight reasoning-based components. For commands requiring interpretation, we adopt a Chain-of-Draft strategy to enable efficient and stable decision-making. To support personalization, we further propose a preference adaptation mechanism that learns from user feedback over time without requiring prompt augmentation or model retraining. We evaluate AdaHome against representative LLM-based baselines under a unified small model setting. AdaHome achieves substantially higher accuracy on direct commands (86.7%) while reducing latency by up to 3$\times$. Furthermore, it maintains competitive performance on ambiguous inputs with lower computational cost. In multi-turn scenarios, AdaHome achieves 88% preference consistency, compared to 52.5% for a prompt augmentation baseline.