@inproceedings{sanger-etal-2021-wbi,
title = "{WBI} at {MEDIQA} 2021: Summarizing Consumer Health Questions with Generative Transformers",
author = {S{\"a}nger, Mario and
Weber, Leon and
Leser, Ulf},
booktitle = "Proceedings of the 20th Workshop on Biomedical Language Processing",
month = jun,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.bionlp-1.9",
doi = "10.18653/v1/2021.bionlp-1.9",
pages = "86--95",
abstract = "This paper describes our contribution for the MEDIQA-2021 Task 1 question summarization competition. We model the task as conditional generation problem. Our concrete pipeline performs a finetuning of the large pretrained generative transformers PEGASUS (Zhang et al.,2020a) and BART (Lewis et al.,2020). We used the resulting models as strong baselines and experimented with (i) integrating structured knowledge via entity embeddings, (ii) ensembling multiple generative models with the generator-discriminator framework and (iii) disentangling summarization and interrogative prediction to achieve further improvements.Our best performing model, a fine-tuned vanilla PEGASUS, reached the second place in the competition with an ROUGE-2-F1 score of 15.99. We observed that all of our additional measures hurt performance (up to 5.2 pp) on the official test set. In course of a post-hoc experimental analysis which uses a larger validation set results indicate slight performance improvements through the proposed extensions. However, further analysis is need to provide stronger evidence.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sanger-etal-2021-wbi">
<titleInfo>
<title>WBI at MEDIQA 2021: Summarizing Consumer Health Questions with Generative Transformers</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mario</namePart>
<namePart type="family">Sänger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leon</namePart>
<namePart type="family">Weber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ulf</namePart>
<namePart type="family">Leser</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-06</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 20th Workshop on Biomedical Language Processing</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper describes our contribution for the MEDIQA-2021 Task 1 question summarization competition. We model the task as conditional generation problem. Our concrete pipeline performs a finetuning of the large pretrained generative transformers PEGASUS (Zhang et al.,2020a) and BART (Lewis et al.,2020). We used the resulting models as strong baselines and experimented with (i) integrating structured knowledge via entity embeddings, (ii) ensembling multiple generative models with the generator-discriminator framework and (iii) disentangling summarization and interrogative prediction to achieve further improvements.Our best performing model, a fine-tuned vanilla PEGASUS, reached the second place in the competition with an ROUGE-2-F1 score of 15.99. We observed that all of our additional measures hurt performance (up to 5.2 pp) on the official test set. In course of a post-hoc experimental analysis which uses a larger validation set results indicate slight performance improvements through the proposed extensions. However, further analysis is need to provide stronger evidence.</abstract>
<identifier type="citekey">sanger-etal-2021-wbi</identifier>
<identifier type="doi">10.18653/v1/2021.bionlp-1.9</identifier>
<location>
<url>https://aclanthology.org/2021.bionlp-1.9</url>
</location>
<part>
<date>2021-06</date>
<extent unit="page">
<start>86</start>
<end>95</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T WBI at MEDIQA 2021: Summarizing Consumer Health Questions with Generative Transformers
%A Sänger, Mario
%A Weber, Leon
%A Leser, Ulf
%S Proceedings of the 20th Workshop on Biomedical Language Processing
%D 2021
%8 June
%I Association for Computational Linguistics
%C Online
%F sanger-etal-2021-wbi
%X This paper describes our contribution for the MEDIQA-2021 Task 1 question summarization competition. We model the task as conditional generation problem. Our concrete pipeline performs a finetuning of the large pretrained generative transformers PEGASUS (Zhang et al.,2020a) and BART (Lewis et al.,2020). We used the resulting models as strong baselines and experimented with (i) integrating structured knowledge via entity embeddings, (ii) ensembling multiple generative models with the generator-discriminator framework and (iii) disentangling summarization and interrogative prediction to achieve further improvements.Our best performing model, a fine-tuned vanilla PEGASUS, reached the second place in the competition with an ROUGE-2-F1 score of 15.99. We observed that all of our additional measures hurt performance (up to 5.2 pp) on the official test set. In course of a post-hoc experimental analysis which uses a larger validation set results indicate slight performance improvements through the proposed extensions. However, further analysis is need to provide stronger evidence.
%R 10.18653/v1/2021.bionlp-1.9
%U https://aclanthology.org/2021.bionlp-1.9
%U https://doi.org/10.18653/v1/2021.bionlp-1.9
%P 86-95
Markdown (Informal)
[WBI at MEDIQA 2021: Summarizing Consumer Health Questions with Generative Transformers](https://aclanthology.org/2021.bionlp-1.9) (Sänger et al., BioNLP 2021)
ACL