arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2407.04050cs.CLcs.AI
收藏

面向基础模型中实体与方面目标情感分析的深度内容理解

Deep Content Understanding Toward Entity and Aspect Target Sentiment Analysis on Foundation Models

IBM Consulting France(IBM咨询法国公司)

Vorakit Vorakitphan, Milos Basic, Guilhaume Leroy Meline

AI总结:

本文提出实体-方面情感三元组抽取(EASTE)任务,通过统一损失方法在 BERT 到 Llama3、Mixtral 等多种模型上结合 LoRA 等适配技术进行探索,并在 SemEval-2016 上取得最先进结果。

摘要:

本文提出实体-方面情感三元组抽取(EASTE),这是一项新颖的基于方面的情感分析(ABSA)任务,它扩展了目标-方面-情感检测(TASD),将方面类别(例如 food#quality)拆分为预定义实体(例如 meal、drink)和方面(例如 taste、freshness),从而增加了更细粒度的复杂性,同时有助于揭示链式方面对其所属实体的真实情感。我们探索了基于 Transformer 架构的语言模型解决 EASTE 任务的能力,从我们提出的统一损失方法出发,通过使用 BERT 架构的 token 分类任务,到文本生成模型如 Flan-T5、Flan-Ul2、Llama2、Llama3 和 Mixtral,并采用不同的对齐技术,如零样本/少样本学习、参数高效微调(PEFT)如低秩适配(LoRA)。模型性能在 SemEval-2016 基准数据集上进行评估,以与现有工作进行公平比较。我们的研究不仅旨在 EASTE 任务上取得高性能,还探讨了模型规模、类型和适配技术对任务性能的影响。最终,我们提供了详细的见解,并在复杂情感分析中取得了最先进的结果。

英文摘要:

Introducing Entity-Aspect Sentiment Triplet Extraction (EASTE), a novel Aspect-Based Sentiment Analysis (ABSA) task which extends Target-Aspect-Sentiment Detection (TASD) by separating aspect categories (e.g., food#quality) into pre-defined entities (e.g., meal, drink) and aspects (e.g., taste, freshness) which add a fine-gainer level of complexity, yet help exposing true sentiment of chained aspect to its entity. We explore the task of EASTE solving capabilities of language models based on transformers architecture from our proposed unified-loss approach via token classification task using BERT architecture to text generative models such as Flan-T5, Flan-Ul2 to Llama2, Llama3 and Mixtral employing different alignment techniques such as zero/few-shot learning, Parameter Efficient Fine Tuning (PEFT) such as Low-Rank Adaptation (LoRA). The model performances are evaluated on the SamEval-2016 benchmark dataset representing the fair comparison to existing works. Our research not only aims to achieve high performance on the EASTE task but also investigates the impact of model size, type, and adaptation techniques on task performance. Ultimately, we provide detailed insights and achieving state-of-the-art results in complex sentiment analysis.

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。

↑