CLAIDec 29, 2024

ELECTRA and GPT-4o: Cost-Effective Partners for Sentiment Analysis

arXiv:2501.00062v211 citationsh-index: 1Proceedings of the 4th International Workshop on Knowledge-Augmented Methods for Natural Language Processing
Originality Incremental advance
AI Analysis

This work addresses cost-effective sentiment analysis for projects with limited resources, presenting incremental improvements through model collaboration and fine-tuning.

This paper tackled sentiment analysis by exploring collaborative approaches between ELECTRA and GPT-4o models, finding that sharing ELECTRA Base fine-tuned predictions with GPT-4o-mini improved performance to 82.50 macro F1 and yielded the lowest cost/performance ratio at $0.12/F1 point, while fine-tuned GPT-4o achieved the highest performance at 86.99 macro F1.

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.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes