用ELECTRA和GPT-4o-mini协作提升情感分析效果，成本更低。
ELECTRA and GPT-4o: Cost-Effective Partners for Sentiment Analysis
- 让ELECTRA预测结果辅助GPT-4o-mini提示，提升分类性能
- 联合模型达82.50宏F1，单位成本仅$0.12/F1点
- 微调后的GPT-4o-mini接近原版性能，成本降低76%
双向Transformer在情感分析中表现优异，大语言模型（LLM）具备零样本学习能力。二者能否协同增效？本文探索ELECTRA与GPT-4o在三分类情感分析中的合作模式。使用来自SST和DynaSent的混合评论数据，对ELECTRA Base/Large及GPT-4o/4o-mini四模型进行微调（FT）。将ELECTRA Base FT的预测标签、概率和检索样例作为输入提供给GPT-4o-mini，显著提升性能（82.50宏F1），优于单一模型（ELECTRA Base FT: 79.14，GPT-4o-mini: 79.41），且成本效益最高（$0.12/F1点）。但当使用微调后的GPT模型时，引入预测反而降低效果。微调版GPT-4o FT表现最佳（86.99宏F1），GPT-4o-mini FT紧随其后（86.70），成本仅为$0.38对比$1.59每F1点。结果表明，以微调编码器输出增强提示是高效提升性能的方法；而微调后的GPT-4o-mini几乎媲美原版，成本低76%，适合资源有限项目。
原文摘要 · Abstract (English)
Bidirectional transformers excel at sentiment analysis, and Large Language Models (LLM) are effective zero-shot learners. Might they perform better as a team? This paper explores collaborative approaches between ELECTRA and GPT-4o for three-way sentiment classification. We fine-tuned (FT) four models (ELECTRA Base/Large, GPT-4o/4o-mini) using a mix of reviews from Stanford Sentiment Treebank (SST) and DynaSent. We provided input from ELECTRA to GPT as: predicted label, probabilities, and retrieved examples. Sharing ELECTRA Base FT predictions with GPT-4o-mini significantly improved performance over either model alone (82.50 macro F1 vs. 79.14 ELECTRA Base FT, 79.41 GPT-4o-mini) and yielded the lowest cost/performance ratio (\$0.12/F1 point). However, when GPT models were fine-tuned, including predictions decreased performance. GPT-4o FT-M was the top performer (86.99), with GPT-4o-mini FT close behind (86.70) at much less cost (\$0.38 vs. \$1.59/F1 point). Our results show that augmenting prompts with predictions from fine-tuned encoders is an efficient way to boost performance, and a fine-tuned GPT-4o-mini is nearly as good as GPT-4o FT at 76% less cost. Both are affordable options for projects with limited resources.
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。