Reproduce “Intelligence of Things: A Spatial Context-Aware Control System for Smart Devices” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Sukanth Kalivarathan, Muhmmad Abrar Raja Mohamed, Aswathy Ravikumar, S Harini
INOT is a spatial context-aware control system for smart devices, leveraging Vision Language Models and Large Language Models to enable natural language commands referencing physical locations. A user study indicated the system reduced cognitive workload by an average of 13.17 points on the NASA-TLX compared to conventional voice assistants.
Key insights
- Integrating VLMs and LLMs allows for dynamic, real-time spatial reasoning about smart environments, enabling natural language control beyond static device identifiers.
- Automated visual recognition and spatial context inference significantly reduce user cognitive load and enhance the intuitiveness of smart home interactions.
- A modular, platform-agnostic system design ensures extensibility and adaptability, allowing for seamless integration with existing IoT ecosystems and broader accessibility.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.