LENS creates over 100,000 sensor-text QA pairs from 258 participants' EMA data and trains a patch-level encoder that projects raw multimodal sensor streams into an LLM's space, enabling generation of clinically grounded depression and anxiety narratives that outperform baselines on NLP and symptom-1
In Proceedings of the Conference on Health, Inference, and Learning (CHIL)
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
PULSE demonstrates that agentic LLM-based investigation of passive smartphone sensing data achieves balanced accuracies of 0.743 (with diary) and 0.713 (sensing-only) for predicting emotion regulation desire and intervention availability in 50 cancer survivors.
Introduces a lightweight gravity-aware routing head that improves macro-F1 on static classes in compressed SensorLLM for human activity recognition on the MHealth dataset.
The survey organizes foundation models for sensor-based HAR into a lifecycle taxonomy and identifies three trajectories: HAR-specific models from scratch, adaptation of general time-series models, and integration with large language models.
citing papers explorer
-
LENS: LLM-Enabled Narrative Synthesis for Mental Health by Aligning Multimodal Sensing with Language Models
LENS creates over 100,000 sensor-text QA pairs from 258 participants' EMA data and trains a patch-level encoder that projects raw multimodal sensor streams into an LLM's space, enabling generation of clinically grounded depression and anxiety narratives that outperform baselines on NLP and symptom-1
-
PULSE: Agentic Investigation with Passive Sensing for Proactive Intervention in Cancer Survivorship
PULSE demonstrates that agentic LLM-based investigation of passive smartphone sensing data achieves balanced accuracies of 0.743 (with diary) and 0.713 (sensing-only) for predicting emotion regulation desire and intervention availability in 50 cancer survivors.
-
Gravity-Aware Hierarchical Routing for Lightweight SensorLLM on Human Activity Recognition
Introduces a lightweight gravity-aware routing head that improves macro-F1 on static classes in compressed SensorLLM for human activity recognition on the MHealth dataset.
-
Foundation Models Defining A New Era In Sensor-based Human Activity Recognition: A Survey And Outlook
The survey organizes foundation models for sensor-based HAR into a lifecycle taxonomy and identifies three trajectories: HAR-specific models from scratch, adaptation of general time-series models, and integration with large language models.