---
title: "Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.15202v1"
permalink: "https://feed.astralor.com/items/itm_57a7b06ca324f7247d302d84"
published: "2026-07-16T16:59:54.000Z"
score: 78
channel: "ai"
tags: ["arXiv","DSM-5-TR","Major Depressive Disorder","Explainable AI","Large Language Models","Self-Evolving Framework"]
---

# Self-Evolving Human-Centered Framework for Explainable Depression Symptom Annotation

## AI commentary

> What stands out in this paper is the integration of self-evolving learning into mental health annotation, with a dual-memory architecture enabling continuous improvement without retraining. For researchers working on medical AI, especially explainability and dataset quality control, this is a highly practical toolkit. If you are building depression-related NLP datasets, the audit trail and expert feedback mechanism are worth adopting.

## At a glance

This paper proposes a self-evolving, expert-in-the-loop annotation framework for building explainable, DSM-5-TR-aligned depression datasets. It operates in three stages: candidate evidence selection, criterion-level analysis, and case-level synthesis. A dual-memory architecture internalizes expert feedback to iteratively improve annotations without retraining. A pilot study shows improved consistency and explainability with reduced manual revision effort.

## Key points

- Proposes a self-evolving MDD annotation framework combining LLM assistance with expert verification for DSM-5-TR-aligned labels. 
- Three-stage pipeline: candidate evidence selection, criterion-level analysis, and case-level synthesis, with export of clinical evidence, reasoning traces, and edit history. 
- Dual-memory architecture (Example Memory and Reflection Memory) internalizes expert feedback for iterative improvement without retraining. 
- Pilot study shows improved annotation consistency and explainability with reduced manual revision effort. 

## Full text

Annotation quality is a major bottleneck in building reliable and explainable artificial intelligence (XAI) systems for mental health research. In depression-related datasets, labels are often assigned without structured evidence, symptom-level justification, or traceable alignment with the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR), limiting both transparency and downstream model interpretability. We propose a self-evolving, expert-in-the-loop annotation framework for Major Depressive Disorder (MDD) that combines large language model (LLM)-assisted labeling with expert verification. The framework is intended to support the construction of explainable, DSM-5-TR-aligned datasets rather than to perform clinical diagnosis. It operates in three stages: candidate evidence selection from textual records, criterion-level DSM-5-TR analysis, and case-level synthesis that produces label-level diagnostic and severity annotations. A dual-memory architecture, composed of Example Memory and Reflection Memory, is designed to internalize expert feedback and iteratively improve future annotations without retraining. We describe this mechanism and leave its evaluation across multiple feedback cycles to future work. In addition to final labels, the framework exports clinical evidence, reasoning traces, and edit histories, enabling comprehensive auditability. In a pilot study using expert-reviewed samples, the proposed approach improves annotation consistency and explainability while reducing manual revision effort.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_57a7b06ca324f7247d302d84