Neck-Learn: Attention-Based Multiple Instance Learning and Ensemble Framework for Ecological Momentary Assessment
Ahsan Jamal · 2026
Read Neck-Learn: Attention-Based Multiple Instance Learning and Ensemble Framework for Ecological Momentary Assessment by Ahsan Jamal (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-02700.txt.
Description
Ahsan Jamal Cheema presents a system for detecting vocal hyperfunction, a voice disorder, from wearable recordings gathered during daily life. Neck-surface sensors can record a week of voice use, but older methods collapse each person's data into one fixed feature set. That discards within-day timing patterns, where fine interactions between voicing traits may signal the disorder.
Cheema's hybrid design keeps both views. Gradient-boosted trees work on day-level summary features, while a convolutional multiple-instance model learns from the day's temporal dynamics directly. On held-out data the system beat the challenge baselines for both pathological and non-pathological hyperfunction, with clear gains in ranking scores. Its placement in the challenge was near the top for both forms. The analysis also surfaces clinically relevant contrasts between the two pathologies.
Catalog record
| CREATOR | Ahsan Jamal |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 26 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.02700