Multi-View Attention Multiple-Instance Learning Enhanced by LLM Reasoning for Cognitive Distortion Detection

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Kim, Jun Seo, Kim, Hyemi, Oh, Woo Joo, Cho, Hongjin, Lee, Hochul, Kim, Hye Hyeon
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866908972087246848
author Kim, Jun Seo
Kim, Hyemi
Oh, Woo Joo
Cho, Hongjin
Lee, Hochul
Kim, Hye Hyeon
author_facet Kim, Jun Seo
Kim, Hyemi
Oh, Woo Joo
Cho, Hongjin
Lee, Hochul
Kim, Hye Hyeon
contents Cognitive distortions have been closely linked to mental health disorders, yet their automatic detection remains challenging due to contextual ambiguity, co-occurrence, and semantic overlap. We propose a novel framework that combines Large Language Models (LLMs) with a Multiple-Instance Learning (MIL) architecture to enhance interpretability and expression-level reasoning. Each utterance is decomposed into Emotion, Logic, and Behavior (ELB) components, which are processed by LLMs to infer multiple distortion instances, each with a predicted type, expression, and model-assigned salience score. These instances are integrated via a Multi-View Gated Attention mechanism for final classification. Experiments on Korean (KoACD) and English (Therapist QA) datasets demonstrate that incorporating ELB and LLM-inferred salience scores improves classification performance, especially for distortions with high interpretive ambiguity. Our results suggest a psychologically grounded and generalizable approach for fine-grained reasoning in mental health NLP. The dataset and implementation details are publicly accessible.
format Preprint
id arxiv_https___arxiv_org_abs_2509_17292
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multi-View Attention Multiple-Instance Learning Enhanced by LLM Reasoning for Cognitive Distortion Detection
Kim, Jun Seo
Kim, Hyemi
Oh, Woo Joo
Cho, Hongjin
Lee, Hochul
Kim, Hye Hyeon
Computation and Language
Artificial Intelligence
Cognitive distortions have been closely linked to mental health disorders, yet their automatic detection remains challenging due to contextual ambiguity, co-occurrence, and semantic overlap. We propose a novel framework that combines Large Language Models (LLMs) with a Multiple-Instance Learning (MIL) architecture to enhance interpretability and expression-level reasoning. Each utterance is decomposed into Emotion, Logic, and Behavior (ELB) components, which are processed by LLMs to infer multiple distortion instances, each with a predicted type, expression, and model-assigned salience score. These instances are integrated via a Multi-View Gated Attention mechanism for final classification. Experiments on Korean (KoACD) and English (Therapist QA) datasets demonstrate that incorporating ELB and LLM-inferred salience scores improves classification performance, especially for distortions with high interpretive ambiguity. Our results suggest a psychologically grounded and generalizable approach for fine-grained reasoning in mental health NLP. The dataset and implementation details are publicly accessible.
title Multi-View Attention Multiple-Instance Learning Enhanced by LLM Reasoning for Cognitive Distortion Detection
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2509.17292