yale-nlp/QTSumm
Viewer • Updated • 7.11k • 54 • 8
How to use yale-nlp/bart-large-finetuned-qtsumm with Transformers:
# Use a pipeline as a high-level helper
# Warning: Pipeline type "summarization" is no longer supported in transformers v5.
# You must load the model directly (see below) or downgrade to v4.x with:
# 'pip install "transformers<5.0.0'
from transformers import pipeline
pipe = pipeline("summarization", model="yale-nlp/bart-large-finetuned-qtsumm") # Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("yale-nlp/bart-large-finetuned-qtsumm")
model = AutoModelForSeq2SeqLM.from_pretrained("yale-nlp/bart-large-finetuned-qtsumm", device_map="auto")QTSumm is a query-focused table summarization dataset proposed in EMNLP 2023 paper QTSUMM: Query-Focused Summarization over Tabular Data. The original Github repository is https://github.com/yale-nlp/QTSumm.
yale-nlp/bart-large-finetuned-qtsumm (based on BART architecture) is initialized with facebook/bart-large and finetuned on the QTSumm dataset.
Check the github repository: https://github.com/yale-nlp/QTSumm
@misc{zhao2023qtsumm,
title={QTSUMM: Query-Focused Summarization over Tabular Data},
author={Yilun Zhao and Zhenting Qi and Linyong Nan and Boyu Mi and Yixin Liu and Weijin Zou and Simeng Han and Xiangru Tang and Yumo Xu and Arman Cohan and Dragomir Radev},
year={2023},
eprint={2305.14303},
archivePrefix={arXiv},
primaryClass={cs.CL}
}