Findings of the Third Automatic Minuting (AutoMin) Challenge

2025-09-17

AI Summary

This work addresses multilingual meeting understanding by introducing a joint task of automatic meeting summarization and transcription-based question answering (QA), targeting English and Czech in two distinct settings: project meetings and European Parliament sessions. Method: We propose a unified evaluation framework supporting structured summary generation (including agenda items, decisions, and action items) and monolingual/cross-lingual QA (e.g., English meeting transcripts with Czech questions)—the first to incorporate meeting transcription QA into multilingual meeting analysis evaluation. Our approach employs large language models (LLMs) as baselines, enhanced with natural language understanding, controllable text generation, and cross-lingual transfer techniques for end-to-end processing. Contribution/Results: Experiments establish the current capabilities and critical limitations of LLMs in multilingual meeting understanding. We release a benchmark dataset, standardized evaluation protocols, and reproducible baselines—enabling systematic advancement in this emerging research area.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Multiagent Systems: Agreement, Argumentation & Negotiation

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

This paper presents the third edition of AutoMin, a shared task on automatic meeting summarization into minutes. In 2025, AutoMin featured the main task of minuting, the creation of structured meeting minutes, as well as a new task: question answering (QA) based on meeting transcripts. The minuting task covered two languages, English and Czech, and two domains: project meetings and European Parliament sessions. The QA task focused solely on project meetings and was available in two settings: monolingual QA in English, and cross-lingual QA, where questions were asked and answered in Czech based on English meetings. Participation in 2025 was more limited compared to previous years, with only one team joining the minuting task and two teams participating in QA. However, as organizers, we included multiple baseline systems to enable a comprehensive evaluation of current (2025) large language models (LLMs) on both tasks.

Problem

• Automatic meeting summarization into structured minutes
• Question answering based on meeting transcripts
• Cross-lingual QA from English meetings to Czech

Innovation

• Automatic meeting summarization into structured minutes
• Question answering based on meeting transcripts
• Cross-lingual QA from English meetings to Czech