# DS@GT at eRisk 2025: From prompts to predictions, benchmarking early depression detection with conversational agent based assessments and temporal attention models

arXiv: https://arxiv.org/abs/2507.10958 · HTML: https://mlsanity.com/paper/2507.10958/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Medical and scientific domain agents, one of 16 directions in LLM agents and tool use.
- Builds on 3 of its 22 references that are in the map.
- First posted on arXiv 15 Jul 2025.

## Placed in

- LLM agents and tool use / [Medical and scientific domain agents](https://mlsanity.com/agents/medical-and-scientific-domain-agents/)

## Builds on (most cited first, 3 of 3)

- [LLM-empowered Chatbots for Psychiatrist and Patient Simulation: Application and Evaluation](https://mlsanity.com/paper/2305.13614/) (2023)
- [PATIENT-Psi: Using Large Language Models to Simulate Patients for Training Mental Health Professionals](https://mlsanity.com/paper/2405.19660/) (2024)
- [MASON-NLP at eRisk 2023: Deep Learning-Based Detection of Depression Symptoms from Social Media Texts](https://mlsanity.com/paper/2310.10941/) (2023)

## Built on by (most cited first, 1 of 1)

- [DS@GT ARC at eRisk 2026: Hybrid Multi-Agent LLM System with Structured Algorithmic Guidance for Conversational Depression Screening](https://mlsanity.com/paper/2607.16712/) (2026)

## Abstract (arXiv)

This Working Note summarizes the participation of the DS@GT team in two eRisk 2025 challenges. For the Pilot Task on conversational depression detection with large language-models (LLMs), we adopted a prompt-engineering strategy in which diverse LLMs conducted BDI-II-based assessments and produced structured JSON outputs. Because ground-truth labels were unavailable, we evaluated cross-model agreement and internal consistency. Our prompt design methodology aligned model outputs with BDI-II criteria and enabled the analysis of conversational cues that influenced the prediction of symptoms. Our best submission, second on the official leaderboard, achieved DCHR = 0.50, ADODL = 0.89, and ASHR = 0.27.