用多任务框架和生成数据提升稀有毒意分类效果
ShriNep@EEUCA 2026: RAKSHAK - Multi-Task DeBERTa with Rationale Distillation and Jigsaw-Augmented Training for Toxic Intent Classification
- 结合理由蒸馏与拼图增强训练，改进DeBERTa模型
- 在极端稀缺数据下仍达0.5883的宏观F1分数
- 适合处理低资源、多类别有毒意图识别任务
本文针对ACL 2026 EEUCA研讨会中的GameTox共享任务，提出两个系统以对《坦克世界》聊天内容进行六类细粒度毒意分类（标签0-5）。严重类别不平衡、领域特异的多语言俚语及罕见类别样本极少（如威胁类仅60样本，极端主义类仅24样本）使该问题极具挑战。主系统RAKSHAK（意为“守护者”）基于DeBERTa-v3-base，融合来自Qwen2.5-14B的理由蒸馏、监督对比损失及专用于罕见类别的二分类头。其训练数据通过跨域迁移自Jigsaw Toxic Comment数据集（16,225样本映射至标签1-4）和100条大模型生成的极端主义样本（标签5）进行增强。次系统M1仅在原始数据上使用焦点损失微调。RAKSHAK在官方测试集上取得0.5883的宏平均F1，排名35支队伍第7；M1为0.5252。消融实验表明，跨域迁移贡献+2.6 F1点，多任务架构再提升+3.7 F1点。
原文摘要 · Abstract (English)
This paper presents two systems for the GameTox Shared Task at the Workshop on EEUCA at ACL 2026, which requires classifying World of Tanks chat utterances into six fine-grained toxic intent categories (Labels 0-5). Severe class imbalance, domain-specific multilingual slang, and extremely scarce data for rare categories such as Threats (Label 4, 60 samples) and Extremism (Label 5, 24 samples) make this a challenging classification problem. Our primary submission, RAKSHAK (rak s. aka, Sanskrit for "Protector"), is a multi-task DeBERTa-v3-base (He et al., 2022) framework combining rationale distillation from Qwen2.5-14B (An et al., 2024), Supervised Contrastive Loss, and dedicated rare-class binary heads. RAKSHAK's training data is augmented with cross-domain transfer from the Jigsaw Toxic Comment dataset (16,225 samples mapped to Labels 1-4) and 100 LLM-generated extremism samples for Label 5. Our secondary system (M1) fine-tunes DeBERTa-v3-base with Focal Loss on the original GameTox data plus the same 100 extremism samples, without Jigsaw transfer. RAKSHAK achieves a Macro F1 of 0.5883 on the official test set, ranking 7th out of 35 participating teams, while M1 achieves 0.5252 Macro F1. An ablation comparing M1 with and without Jigsaw data shows that cross-domain transfer accounts for +2.6 F1 points, while RAKSHAK's multi-task architecture contributes a further +3.7 points.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。