@inproceedings{orbach-goldberg-2020-facts2story,
title = "{F}acts2{S}tory: Controlling Text Generation by Key Facts",
author = "Orbach, Eyal and
Goldberg, Yoav",
booktitle = "Proceedings of the 28th International Conference on Computational Linguistics",
month = dec,
year = "2020",
address = "Barcelona, Spain (Online)",
publisher = "International Committee on Computational Linguistics",
url = "https://aclanthology.org/2020.coling-main.211",
doi = "10.18653/v1/2020.coling-main.211",
pages = "2329--2345",
abstract = "Recent advancements in self-attention neural network architectures have raised the bar for open-ended text generation. Yet, while current methods are capable of producing a coherent text which is several hundred words long, attaining control over the content that is being generated{---}as well as evaluating it{---}are still open questions. We propose a controlled generation task which is based on expanding a sequence of facts, expressed in natural language, into a longer narrative. We introduce human-based evaluation metrics for this task, as well as a method for deriving a large training dataset. We evaluate three methods on this task, based on fine-tuning pre-trained models. We show that while auto-regressive, unidirectional Language Models such as GPT2 produce better fluency, they struggle to adhere to the requested facts. We propose a plan-and-cloze model (using fine-tuned XLNet) which produces competitive fluency while adhering to the requested content.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="orbach-goldberg-2020-facts2story">
<titleInfo>
<title>Facts2Story: Controlling Text Generation by Key Facts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Eyal</namePart>
<namePart type="family">Orbach</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-dec</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 28th International Conference on Computational Linguistics</title>
</titleInfo>
<originInfo>
<publisher>International Committee on Computational Linguistics</publisher>
<place>
<placeTerm type="text">Barcelona, Spain (Online)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recent advancements in self-attention neural network architectures have raised the bar for open-ended text generation. Yet, while current methods are capable of producing a coherent text which is several hundred words long, attaining control over the content that is being generated—as well as evaluating it—are still open questions. We propose a controlled generation task which is based on expanding a sequence of facts, expressed in natural language, into a longer narrative. We introduce human-based evaluation metrics for this task, as well as a method for deriving a large training dataset. We evaluate three methods on this task, based on fine-tuning pre-trained models. We show that while auto-regressive, unidirectional Language Models such as GPT2 produce better fluency, they struggle to adhere to the requested facts. We propose a plan-and-cloze model (using fine-tuned XLNet) which produces competitive fluency while adhering to the requested content.</abstract>
<identifier type="citekey">orbach-goldberg-2020-facts2story</identifier>
<identifier type="doi">10.18653/v1/2020.coling-main.211</identifier>
<location>
<url>https://aclanthology.org/2020.coling-main.211</url>
</location>
<part>
<date>2020-dec</date>
<extent unit="page">
<start>2329</start>
<end>2345</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Facts2Story: Controlling Text Generation by Key Facts
%A Orbach, Eyal
%A Goldberg, Yoav
%S Proceedings of the 28th International Conference on Computational Linguistics
%D 2020
%8 dec
%I International Committee on Computational Linguistics
%C Barcelona, Spain (Online)
%F orbach-goldberg-2020-facts2story
%X Recent advancements in self-attention neural network architectures have raised the bar for open-ended text generation. Yet, while current methods are capable of producing a coherent text which is several hundred words long, attaining control over the content that is being generated—as well as evaluating it—are still open questions. We propose a controlled generation task which is based on expanding a sequence of facts, expressed in natural language, into a longer narrative. We introduce human-based evaluation metrics for this task, as well as a method for deriving a large training dataset. We evaluate three methods on this task, based on fine-tuning pre-trained models. We show that while auto-regressive, unidirectional Language Models such as GPT2 produce better fluency, they struggle to adhere to the requested facts. We propose a plan-and-cloze model (using fine-tuned XLNet) which produces competitive fluency while adhering to the requested content.
%R 10.18653/v1/2020.coling-main.211
%U https://aclanthology.org/2020.coling-main.211
%U https://doi.org/10.18653/v1/2020.coling-main.211
%P 2329-2345
Markdown (Informal)
[Facts2Story: Controlling Text Generation by Key Facts](https://aclanthology.org/2020.coling-main.211) (Orbach & Goldberg, COLING 2020)
ACL
- Eyal Orbach and Yoav Goldberg. 2020. Facts2Story: Controlling Text Generation by Key Facts. In Proceedings of the 28th International Conference on Computational Linguistics, pages 2329–2345, Barcelona, Spain (Online). International Committee on Computational Linguistics.