Generate Then Correct: Single Shot Global Correction for Aspect Sentiment Quad Prediction

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: He, Shidong, Wang, Haoyu, Luo, Wenjie
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910102262382592
author He, Shidong
Wang, Haoyu
Luo, Wenjie
author_facet He, Shidong
Wang, Haoyu
Luo, Wenjie
contents Aspect-based sentiment analysis (ABSA) extracts aspect-level sentiment signals from user-generated text, supports product analytics, experience monitoring, and public-opinion tracking, and is central to fine-grained opinion mining. A key challenge in ABSA is aspect sentiment quad prediction (ASQP), which requires identifying four elements: the aspect term, the aspect category, the opinion term, and the sentiment polarity. However, existing studies usually linearize the unordered quad set into a fixed-order template and decode it left-to-right. With teacher forcing training, the resulting training-inference mismatch (exposure bias) lets early prefix errors propagate to later elements. The linearization order determines which elements appear earlier in the prefix, so this propagation becomes order-sensitive and is hard to repair in a single pass. To address this, we propose a method, Generate-then-Correct (G2C): a generator drafts quads and a corrector performs a single-shot, sequence-level global correction trained on LLM-synthesized drafts with common error patterns. On the Rest15 and Rest16 datasets, G2C outperforms strong baseline models.
format Preprint
id arxiv_https___arxiv_org_abs_2603_13777
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Generate Then Correct: Single Shot Global Correction for Aspect Sentiment Quad Prediction
He, Shidong
Wang, Haoyu
Luo, Wenjie
Computation and Language
Artificial Intelligence
Aspect-based sentiment analysis (ABSA) extracts aspect-level sentiment signals from user-generated text, supports product analytics, experience monitoring, and public-opinion tracking, and is central to fine-grained opinion mining. A key challenge in ABSA is aspect sentiment quad prediction (ASQP), which requires identifying four elements: the aspect term, the aspect category, the opinion term, and the sentiment polarity. However, existing studies usually linearize the unordered quad set into a fixed-order template and decode it left-to-right. With teacher forcing training, the resulting training-inference mismatch (exposure bias) lets early prefix errors propagate to later elements. The linearization order determines which elements appear earlier in the prefix, so this propagation becomes order-sensitive and is hard to repair in a single pass. To address this, we propose a method, Generate-then-Correct (G2C): a generator drafts quads and a corrector performs a single-shot, sequence-level global correction trained on LLM-synthesized drafts with common error patterns. On the Rest15 and Rest16 datasets, G2C outperforms strong baseline models.
title Generate Then Correct: Single Shot Global Correction for Aspect Sentiment Quad Prediction
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2603.13777