TRACE: Temporal Reasoning over Context and Evidence for Activity Recognition in Smart Homes
On the map: LLM reasoning → Spatial and embodied reasoning
Abstract (arXiv)
Human activity recognition (HAR) in smart homes remains challenging because many daily activities exhibit similar local sensor patterns, while minimally intrusive sensing provides sparse and ambiguous observations. As a result, methods based on short temporal or event windows often fail to capture the broader temporal and behavioral context needed for reliable activity understanding. We present TRACE (Temporal Reasoning over Context and Evidence), a contextual activity recognition framework for smart homes that integrates multi-source sensor evidence with user-specific contextual priors to improve activity interpretation. Rather than treating recognition as a local classification problem, TRACE leverages contextual reasoning to resolve ambiguities, reduce fragmented predictions, and infer more semantically specific activities. We evaluate TRACE on public benchmarks and in a deployment study conducted in our smart-home environment. Results show that TRACE improves recognition accuracy for semantically complex activities, produces more temporally coherent predictions that better align with user-specific routines, and maintains robust performance under cross-domain transfer and missing-modality conditions. These findings demonstrate the value of contextual reasoning for advancing smart-home HAR.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 13 of its 102 references that are in the map. First posted on arXiv 4 May 2026.
Builds on
13 of 102 references.
References · 13
- 2025Llasa: Scaling Train-Time and Inference-Time Compute for Llama-based Speech Synthesiscited by 70
- 2024HARGPT: Are LLMs Zero-Shot Human Activity Recognizers?cited by 54
- 2022A Survey of Deep Causal Models and Their Industrial Applicationscited by 10
- 2025RAG-HAR: Retrieval Augmented Generation-based Human Activity Recognitioncited by 7
- 2024Large Language Model-Guided Semantic Alignment for Human Activity Recognitioncited by 6
- 2024Bridging the gap between natural user expression with complex automation programming in smart homescited by 5
- 2025Using LLMs for Late Multimodal Sensor Fusion for Activity Recognitioncited by 5
- 2023Know Thy Neighbors: A Graph Based Approach for Effective Sensor-Based Human Activity Recognition in Smart Homescited by 2
- 2026Knowledge Distillation for LLM-Based Human Activity Recognition in Homescited by 2
- 2026Improving Zero-shot ADL Recognition with Large Language Models through Event-based Context and Confidencecited by 2
- 2024Multidimensional Human Activity Recognition With Large Language Model: A Conceptual Frameworkcited by 1
- 2025On-device Large Multi-modal Agent for Human Activity Recognitioncited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2026Evaluating Few-Shot Temporal Reasoning of LLMs for Human Activity Prediction in Smart Environments
- 2026TRACE-TS: Attribution-Grounded and Traceable Sensor-Language Reasoning for Human Activity Understanding
- 2025MuRAL: A Multi-Resident Ambient Sensor Dataset Annotated with Natural Language for Activities of Daily Living
- 2026CHEAR: Confusion-Hierarchy Evidence-Aware Reasoning for Cost-Efficient Human Activity Recognition
- 2026OrganicHAR: Towards Activity Discovery in Organic Settings for Privacy Preserving Sensors Using Efficient Video Analysis
- 2026SMH-Bench: Benchmarking LLM Agents for Environment-Grounded Reasoning and Action in Smart Homes
- 2025ZARA: Training-Free Motion Time-Series Reasoning via Evidence-Grounded LLM Agents
- 2024SensorLLM: Aligning Large Language Models with Motion Sensors for Human Activity Recognition