Placeholder chapter page. Sections are produced by the book-skills pipeline.
Sections
- 21.1 Aligning sensor representations with language
- 21.2 Sensor-to-text: captioning and describing signals
- 21.3 Sensor question answering (SensorQA, SensorBench, OpenSQA)
- 21.4 LLM-based reasoning over sensor streams (LLaSA, SensorLLM, HARGPT, LLMSense)
- 21.5 Prompting and retrieval over sensor context
- 21.6 Grounding, hallucination, and verification for physical signals
- 21.7 Evaluation of sensor-language systems
Lab 21
build a sensor-QA prototype that answers natural-language questions over an activity/telemetry stream, with a grounding check.