Abstract. In-game toxic language has emerged as a critical concern in the gaming industry and community. While several frameworks and models for online game toxicity analysis have been proposed, detecting toxicity in player chat utterances remains a formidable challenge: stemming not only from the extremely short length of such utterances but also from the heavy reliance on game slang, abbreviations, and domain-specific jargon, which generic language models are poorly suited to recognize. This project introduces the best-preforming model for the in-game toxic language detection from the real-world in-game chat data.

BRAR (Bi-directional Representations with Attention Residuals) integrates:

• BiLSTM performs feature extraction on the input data in sequence.
• Attention Residual aims to understand the global information and find the main ideas of the input utterance.
• Label Forcing enhances the feature representation to form the emission scores of the CRF layer.
• CRF enforces valid tag transitions and prevents locally optimal but globally inconsistent predictions.

The CONDA dataset used in this project comprises 44,869 player chat utterances, derived from chat logs of 1,921 Dota 2 matches and annotated with the following six distinct slot labels:

• T: Toxicity
• C: Character
• D: Dota-specific
• S: Game Slang
• P: Pronoun
• O: Other

python src/brar.py

toxic-language-detection/
├── data/              # CONDA dataset
│   ├── train.csv
│   ├── val.csv
│   └── CONDA_test_original.csv
├── image/             # Model architecture diagram
│   └── brar.png
├── src/
│   └── brar.py        # BRAR model implementation
├── requirements.txt   # Python dependencies
├── .gitignore
└── README.md