用提示编码器+集成学习，多语言情绪识别效果更优
University of Indonesia at SemEval-2025 Task 11: Evaluating State-of-the-Art Encoders for Multi-Label Emotion Detection
- 用提示编码器搭配分类器，不全量微调模型
- 最佳模型在28种语言上平均F1-macro达56.58
- 适合多语言情绪分析任务的高效方案
本文介绍了我们在SemEval 2025 Task 11 Track A中的方法，聚焦于28种语言的多标签情绪分类。我们探索了两种主要策略：全量微调Transformer模型与仅训练分类器，并评估了不同微调方式、模型架构、损失函数、编码器和分类器的组合。研究发现，在mE5和BGE等提示编码器上训练分类器，性能显著优于对XLM-RoBERTa和mBERT进行全量微调。我们最终在排行榜上表现最好的模型是一个结合多个BGE模型的集成系统，使用CatBoost作为分类器，配置各异。该集成模型在所有语言上的平均F1-macro得分为56.58。
原文摘要 · Abstract (English)
This paper presents our approach for SemEval 2025 Task 11 Track A, focusing on multilabel emotion classification across 28 languages. We explore two main strategies: fully fine-tuning transformer models and classifier-only training, evaluating different settings such as fine-tuning strategies, model architectures, loss functions, encoders, and classifiers. Our findings suggest that training a classifier on top of prompt-based encoders such as mE5 and BGE yields significantly better results than fully fine-tuning XLMR and mBERT. Our best-performing model on the final leaderboard is an ensemble combining multiple BGE models, where CatBoost serves as the classifier, with different configurations. This ensemble achieves an average F1-macro score of 56.58 across all languages.
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。