AutoMin 2025: Meeting Summarization and QA Challenge

AutoMin 2025, part of SIGDIAL 2025, challenges researchers to summarize meetings and answer questions using LLMs. Register from March 1st, 2025.

The AutoMin 2025 challenge, part of SIGDIAL 2025, invites researchers to participate in the third bi-annual shared task on meeting summarization and question answering. This challenge aims to assess the capabilities of large language models (LLMs) in summarizing and retrieving information from long contexts, such as hour-long meetings.

The tasks include:

• Task A (Minuting): Automatically creating structured meeting minutes from multiparty transcripts, evaluated via both automatic and manual metrics.
• Task B (Question Answering): Answering questions based on meeting transcripts, either monolingually (English questions on English transcripts) or cross-lingually (Czech questions on English transcripts).

Participants can register from March 1st, 2025, and the challenge will culminate in a workshop on August 28th, 2025. For more details, visit the challenge website.

Tags: AutoMin 2025, SIGDIAL 2025, meeting summarization, question answering, LLMs, NLP