通过动态检索与多轮投票，提升中文仇恨言论细粒度识别准确率
System Report for CCL25-Eval Task 10: SRAG-MAV for Fine-Grained Chinese Hate Speech Recognition
- 将四元组提取转为三元组，结合动态检索生成上下文提示
- 在ToxiCN数据集上平均得分37.505，显著优于GPT-4o和微调版Qwen2.5-7B
- 适合需要高精度中文敏感内容识别的系统开发者参考
本文介绍我们在CCL25-Eval Task 10中的系统方案，针对细粒度中文仇恨言论识别（FGCHSR）问题。提出SRAG-MAV框架，融合任务重构（TR）、自检索增强生成（SRAG）与多轮累积投票（MAV）。方法将四元组提取任务重构为三元组提取，利用训练集动态检索生成上下文提示，并通过多轮推理与投票提升输出稳定性与性能。基于Qwen2.5-7B模型，该系统在STATE ToxiCN数据集上取得硬评分26.66、软评分48.35、平均分37.505，显著优于GPT-4o（平均分15.63）和微调版Qwen2.5-7B（平均分35.365）。代码已开源：https://github.com/king-wang123/CCL25-SRAG-MAV。
原文摘要 · Abstract (English)
This paper presents our system for CCL25-Eval Task 10, addressing Fine-Grained Chinese Hate Speech Recognition (FGCHSR). We propose a novel SRAG-MAV framework that synergistically integrates task reformulation(TR), Self-Retrieval-Augmented Generation (SRAG), and Multi-Round Accumulative Voting (MAV). Our method reformulates the quadruplet extraction task into triplet extraction, uses dynamic retrieval from the training set to create contextual prompts, and applies multi-round inference with voting to improve output stability and performance. Our system, based on the Qwen2.5-7B model, achieves a Hard Score of 26.66, a Soft Score of 48.35, and an Average Score of 37.505 on the STATE ToxiCN dataset, significantly outperforming baselines such as GPT-4o (Average Score 15.63) and fine-tuned Qwen2.5-7B (Average Score 35.365). The code is available at https://github.com/king-wang123/CCL25-SRAG-MAV.
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。