arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2408.09629cs.CL
收藏

一种结合第一代Transformer与开源大语言模型的自动文本分类策略

A Strategy to Combine 1stGen Transformers and Open LLMs for Automatic Text Classification

Federal University of Minas Gerais(米纳斯吉拉斯联邦大学);Federal University of São João Del-Rei(圣若昂德尔雷伊联邦大学)

Claudio M. V. de Andrade, Washington Cunha, Davi Reis, Adriana Silvina Pagano, Leonardo Rocha, Marcos André Gonçalves

AI总结:

本研究对比第一代Transformer与开源大语言模型在情感分析任务上的性能,提出基于置信度的融合策略,用低成本实现接近微调大模型的分类效果。

摘要:

Transformer模型已取得当前最优结果,作为第一代Transformer(1stTR)演进产物的大语言模型(LLM)被视为多项自然语言处理(NLP)任务的前沿技术。然而,现有文献尚未最终证实LLM在所有NLP任务中都能持续优于1stTR。本研究在11个情感分析数据集上,对比了3种1stTR(BERT、RoBERTa和BART)与2种开源LLM(Llama 2和Bloom)的性能。结果表明,开源LLM仅在微调的情况下,能在11个数据集中的8个上适度优于或持平于1stTR。鉴于仅获得适度提升却需付出高昂成本,这类模型在成本敏感场景中的实际适用性存疑。在此背景下,本文提出一种基于置信度的策略,可根据预测确定性无缝整合1stTR与开源LLM:高置信度文档由成本效益更高的1stTR分类,不确定的案例则由LLM以零样本或少样本模式处理,成本远低于微调版本。情感分析实验表明,该方案不仅优于1stTR、零样本及少样本LLM,还能以极低的成本达到与微调LLM相近的性能。

英文摘要:

Transformer models have achieved state-of-the-art results, with Large Language Models (LLMs), an evolution of first-generation transformers (1stTR), being considered the cutting edge in several NLP tasks. However, the literature has yet to conclusively demonstrate that LLMs consistently outperform 1stTRs across all NLP tasks. This study compares three 1stTRs (BERT, RoBERTa, and BART) with two open LLMs (Llama 2 and Bloom) across 11 sentiment analysis datasets. The results indicate that open LLMs may moderately outperform or match 1stTRs in 8 out of 11 datasets but only when fine-tuned. Given this substantial cost for only moderate gains, the practical applicability of these models in cost-sensitive scenarios is questionable. In this context, a confidence-based strategy that seamlessly integrates 1stTRs with open LLMs based on prediction certainty is proposed. High-confidence documents are classified by the more cost-effective 1stTRs, while uncertain cases are handled by LLMs in zero-shot or few-shot modes, at a much lower cost than fine-tuned versions. Experiments in sentiment analysis demonstrate that our solution not only outperforms 1stTRs, zero-shot, and few-shot LLMs but also competes closely with fine-tuned LLMs at a fraction of the cost.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑