Trail 框架用层级类型化决策实现本地智能家居控制
Trail: Accurate Local Smart Home Control Using Decision Models
智能家居命令不想走云端？这篇论文把控制拆成房间、设备、操作三层选择，全程本地跑，还带多轮记忆，准确率数据挺扎实。
arXiv 论文提出 Trail，一个完全本地的智能家居控制框架，用层级类型化决策取代开放式命令生成，从运行时设备注册表中依次选择房间、设备和操作。该框架在无法可靠解析请求时可基于置信度选择拒答，并配有结构化记忆模块支持用户偏好和多轮交互。作者基于 HomeBench 生成 Trek-Base、Trek-Robust、Trek-Memory 三个数据集做评估。Trail 在 Trek-Robust 上达到 85.3% 准确率，超过微调后的 Qwen3.5-4B 基线的 69.8%；在依赖前文轮次的请求上解决率达 75.6%，对比基线的 42.5%。
Trail: Accurate Local Smart Home Control Using Decision Models
Natural-language driven smart home control remains challenging: cloud-hosted Large Language Models (LLMs) raise privacy concerns, while local Small Language Models (SLMs) can generate invalid actions or misinterpret requests. We introduce Trail (Typed Registry-Anchored Inference for Language commands in Smart Homes), a fully local framework that replaces open-ended command generation with hierarchical typed decisions. Trail exploits the natural structure of a home's action space, selecting a room, device, and operation from candidates derived from the runtime device registry. This ensures valid actions while enabling confidence-based abstention when requests cannot be reliably resolved. We also design a structured memory module that uses typed decisions to support user preferences, device references, and multi-turn interactions. To evaluate Trail, we extend HomeBench and generate three datasets: Trek-Base, which tests natural-language variation; Trek-Robust, which evaluates robustness to changes in the home and its description; and Trek-Memory, which assesses conversational memory across multi-turn interactions. Trail achieves 85.3% accuracy on Trek-Robust versus 69.8% for a finetuned Qwen3.5-4B baseline, while remaining competitive on Trek-Base requests. With memory, Trail resolves 75.6% of the requests that depend on earlier turns, compared with 42.5% for the finetuned baseline given conversation history.