# HiPAD: A Unified Gait, Audio-Visual, and Personality Fusion Network for Depression Detection, with a Hierarchical Personality Ablation and the Calibration Limits of Domain-Adapted LLMs

arXiv: https://arxiv.org/abs/2610.14961 · HTML: https://mlsanity.com/paper/2610.14961/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 38 references that are in the map.
- First posted on arXiv 10 Oct 2026.

## Builds on (most cited first, 10 of 10)

- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [Depression and Self-Harm Risk Assessment in Online Forums](https://mlsanity.com/paper/1709.01848/) (2017)
- [On the Calibration of Large Language Models and Alignment](https://mlsanity.com/paper/2311.13240/) (2023)
- [Automated speech-based screening of depression using deep convolutional neural networks](https://mlsanity.com/paper/1912.01115/) (2019)
- [The First MPDD Challenge: Multimodal Personality-aware Depression Detection](https://mlsanity.com/paper/2505.10034/) (2025)
- [Automatic Depression Assessment using Machine Learning: A Comprehensive Survey](https://mlsanity.com/paper/2506.18915/) (2025)
- [Personality-guided Public-Private Domain Disentangled Hypergraph-Former Network for Multimodal Depression Detection](https://mlsanity.com/paper/2511.12460/) (2025)
- [Personality-Enhanced Multimodal Depression Detection in the Elderly](https://mlsanity.com/paper/2510.08004/) (2025)

## Abstract (arXiv)

We present HiPAD, a single fusion network submitted to all six sub-tracks of the MPDD-AVG 2026 depression-detection challenge, covering the Elder and Young cohorts and three modality sets (gait, audio-visual, and both), each combined with personality descriptions. The network encodes each available modality with a shared BiLSTM, zero-pads absent modalities, and applies cross-modal attention only when at least two of gait, audio and video are present. Personality, which dominates our results, is kept out of the attention and fused by concatenation. Because the architecture is identical in every sub-track, differences between them reflect the data rather than the model, and we use this to study three questions. Our main finding is that the personality description, roughly 150 tokens about who the subject is, carries more depression signal than audio, video, and gait combined: gait plus personality alone (TLS 0.647) beats every configuration that includes audio and video. First, a four-level ablation of the personality description (L0-L3) quantifies this: removing every personality-derived feature, including the Big Five scores that earlier MPDD ablations keep, lowers the Track-Level Score by 0.499 to 0.507 and leaves the model at chance. Second, domain-adapting the personality encoder with LoRA (7B) or QLoRA (72B) on PHQ-8/9 data raises binary Cohen's kappa by 0.13 but lowers the overall score by distorting the PHQ-9 regression; a frozen RoBERTa-large encoder does better at this sample size. Third, at fixed architecture, handcrafted clinical features (eGeMAPS, OpenFace action units) outperform generic deep features (wav2vec2, DenseNet) on the audio-visual tracks. The unified system reaches a cross-track G-Score of 0.482; with under 100 subjects per cohort, the limiting factors are the size of the representation and the calibration of the outputs, not the choice of architecture.