QMSum is a new human-annotated benchmark for query-based multi-domain meeting summarization task, which consists of 1,808 query-summary pairs over 232 meetings ...

QMSum is a new human-annotated benchmark for query-based multi-domain meeting summarisation task, which consists of 1,808 query-summary pairs over 232 ...

13 апр. 2021 г. · QMSum consists of 1,808 query-summary pairs over 232 meetings in multiple domains. Besides, we investigate a locate-then-summarize method and ...

qmsum-cleaned prefixes. It's worth noting that each "document" in input is prefixed by a question/prompt on what the model is supposed to do.

We are releasing our dataset and baselines to support additional research in query- focused meeting summarization. Overall, our contributions are listed as ...

Dataset for NAACL 2021 paper: "QMSum: A New Benchmark for Query-based Multi-domain Meeting Summarization" - QMSum/data_process.ipynb at main ...

We're on a journey to advance and democratize artificial intelligence through open source and open science.

This work defines a new query-based multi-domain meeting summarization task, where models have to select and summarize relevant spans of meetings in response ...

The QMSum [28] dataset aggregates three public data sources (ICSI, AMI, and Parliament meetings from Welsh and Canada) and generates minutes for the text ...

To enhance the explainability of meeting summarization, we construct a new dataset called “ExplainMeetSum,” an augmented version of QMSum, by newly annotating ...