Fine-grained Verbal Attack Detection via a Hierarchical Divide-and-Conquer Framework

TL;DR

This paper introduces a hierarchical framework and dataset for fine-grained verbal attack detection in Chinese social media, emphasizing structured modeling of conversational context to improve detection accuracy.

Contribution

It presents a novel hierarchical dataset and a divide-and-conquer framework that decomposes attack detection into subtasks, enhancing detection of implicit and context-dependent verbal attacks.

Findings

Smaller models with the hierarchical framework outperform larger monolithic models.

The dataset captures complex reply structures and chronological order.

Structured task decomposition improves attack detection accuracy.

Abstract

In the digital era, effective identification and analysis of verbal attacks are essential for maintaining online civility and ensuring social security. However, existing research is limited by insufficient modeling of conversational structure and contextual dependency, particularly in Chinese social media where implicit attacks are prevalent. Current attack detection studies often emphasize general semantic understanding while overlooking user response relationships, hindering the identification of implicit and context-dependent attacks. To address these challenges, we present the novel "Hierarchical Attack Comment Detection" dataset and propose a divide-and-conquer, fine-grained framework for verbal attack recognition based on spatiotemporal information. The proposed dataset explicitly encodes hierarchical reply structures and chronological order, capturing complex interaction patterns in multi-turn discussions. Building on this dataset, the framework decomposes attack detection into hierarchical subtasks, where specialized lightweight models handle explicit detection, implicit intent inference, and target identification under constrained context. Extensive experiments on the proposed dataset and benchmark intention detection datasets show that smaller models using our framework significantly outperform larger monolithic models relying on parameter scaling, demonstrating the effectiveness of structured task decomposition.