# JiraiBench: A Bilingual Benchmark for Evaluating Large Language Models' Detection of Human Self-Destructive Behavior Content in Jirai Community

arXiv: https://arxiv.org/abs/2503.21679 · HTML: https://mlsanity.com/paper/2503.21679/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 45 references that are in the map.
- First posted on arXiv 27 Mar 2025.

## Builds on (most cited first, 8 of 8)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Do Llamas Work in English? On the Latent Language of Multilingual Transformers](https://mlsanity.com/paper/2402.10588/) (2024)
- [MentalBERT: Publicly Available Pretrained Language Models for Mental Healthcare](https://mlsanity.com/paper/2110.15621/) (2021)
- [SeqZero: Few-shot Compositional Semantic Parsing with Sequential Prompts and Zero-shot Models](https://mlsanity.com/paper/2205.07381/) (2022)
- [ToxiCloakCN: Evaluating Robustness of Offensive Language Detection in Chinese with Cloaking Perturbations](https://mlsanity.com/paper/2406.12223/) (2024)
- [Identifying Depressive Symptoms from Tweets: Figurative Language Enabled Multitask Learning Framework](https://mlsanity.com/paper/2011.06149/) (2020)
- [Language and Mental Health: Measures of Emotion Dynamics from Text as Linguistic Biosocial Markers](https://mlsanity.com/paper/2310.17369/) (2023)
- [Detection and Classification of mental illnesses on social media using RoBERTa](https://mlsanity.com/paper/2011.11226/) (2020)

## Abstract (arXiv)

This paper introduces JiraiBench, the first bilingual benchmark for evaluating large language models' effectiveness in detecting self-destructive content across Chinese and Japanese social media communities. Focusing on the transnational "Jirai" (landmine) online subculture that encompasses multiple forms of self-destructive behaviors including drug overdose, eating disorders, and self-harm, we present a comprehensive evaluation framework incorporating both linguistic and cultural dimensions. Our dataset comprises 10,419 Chinese posts and 5,000 Japanese posts with multidimensional annotation along three behavioral categories, achieving substantial inter-annotator agreement. Experimental evaluations across four state-of-the-art models reveal significant performance variations based on instructional language, with Japanese prompts unexpectedly outperforming Chinese prompts when processing Chinese content. This emergent cross-cultural transfer suggests that cultural proximity can sometimes outweigh linguistic similarity in detection tasks. Cross-lingual transfer experiments with fine-tuned models further demonstrate the potential for knowledge transfer between these language systems without explicit target language training. These findings highlight the need for culturally-informed approaches to multilingual content moderation and provide empirical evidence for the importance of cultural context in developing more effective detection systems for vulnerable online communities.