Financial RAG Answer Quality Regression with Table-Text Evidence Features on RAGBench FinQA and eManual

Financial retrieval-augmented generation (RAG) systems must answer questions that combine narrative disclosures, tabular values, percentages, fiscal periods, and arithmetic relations. This study formulates answer evaluation as continuous quality regression and binary adherence classification on the...

Täydet tiedot

Bibliografiset tiedot

Päätekijä:	Lewis MacDonald
Aineistotyyppi:	Artigo
Kieli:	Inglês
Julkaistu:	Scitific Publication Center 2026-01-01
Sarja:	Artificial Intelligence and Machine Learning Review
Aiheet:	retrieval-augmented generation financial question answering FinQA RAGBench answer-quality regression table-text evidence
Linkit:	https://scipublication.com/index.php/AIMLR/article/view/418
Tagit:	Lisää tagi Ei tageja, Lisää ensimmäinen tagi!