RV-HATE: Reinforced Multi-Module Voting for Implicit Hate Speech Detection

2025-10-12

AI Summary

Implicit hate speech detection faces challenges from heterogeneous data sources, linguistically complex features, and the lack of adaptability in existing methods. Method: This paper proposes a dynamic detection framework guided by data characteristics, employing a multi-module collaborative architecture to jointly model surface-level linguistic, semantically implicit, and context-dependent features. It integrates reinforcement learning to dynamically optimize module weights and adopts an interpretable ensemble voting mechanism to enhance cross-dataset generalization. Contribution/Results: Extensive experiments on multiple benchmark hate speech datasets demonstrate that the proposed method significantly outperforms static baseline models, achieving an average accuracy improvement of 4.2%. Moreover, it enables interpretable, data-characteristic-driven decision analysis, establishing a novel paradigm for robust hate speech identification over heterogeneous data.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Machine Learning: Multimodal Learning
• Humans and AI: Human-in-the-loop Machine Learning

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Hate speech remains prevalent in human society and continues to evolve in its forms and expressions. Modern advancements in internet and online anonymity accelerate its rapid spread and complicate its detection. However, hate speech datasets exhibit diverse characteristics primarily because they are constructed from different sources and platforms, each reflecting different linguistic styles and social contexts. Despite this diversity, prior studies on hate speech detection often rely on fixed methodologies without adapting to data-specific features. We introduce RV-HATE, a detection framework designed to account for the dataset-specific characteristics of each hate speech dataset. RV-HATE consists of multiple specialized modules, where each module focuses on distinct linguistic or contextual features of hate speech. The framework employs reinforcement learning to optimize weights that determine the contribution of each module for a given dataset. A voting mechanism then aggregates the module outputs to produce the final decision. RV-HATE offers two primary advantages: (1)~it improves detection accuracy by tailoring the detection process to dataset-specific attributes, and (2)~it also provides interpretable insights into the distinctive features of each dataset. Consequently, our approach effectively addresses implicit hate speech and achieves superior performance compared to conventional static methods. Our code is available at https://github.com/leeyejin1231/RV-HATE.

Problem

• Detecting implicit hate speech across diverse datasets
• Adapting detection methods to dataset-specific linguistic features
• Improving accuracy with multi-module reinforcement learning voting

Innovation

• Multi-module voting framework for hate speech detection
• Reinforcement learning optimizes dataset-specific module weights
• Voting mechanism aggregates outputs for final decision