GETALP@AutoMin 2025: Leveraging RAG to Answer Questions based on Meeting Transcripts

TL;DR

This paper presents GETALP's RAG-based system with AMR for question-answering on meeting transcripts, achieving high-quality responses for 35% of questions and improving participant distinction.

Contribution

It introduces three novel systems combining RAG and AMR for meeting transcript question-answering, demonstrating the effectiveness of AMR in this context.

Findings

AMR integration improves answer quality for 35% of questions.

Enhanced ability to distinguish between participants in answers.

RAG with AMR outperforms baseline methods.

Abstract

This paper documents GETALP's submission to the Third Run of the Automatic Minuting Shared Task at SIGDial 2025. We participated in Task B: question-answering based on meeting transcripts. Our method is based on a retrieval augmented generation (RAG) system and Abstract Meaning Representations (AMR). We propose three systems combining these two approaches. Our results show that incorporating AMR leads to high-quality responses for approximately 35% of the questions and provides notable improvements in answering questions that involve distinguishing between different participants (e.g., who questions).