@inproceedings{sancheti-rudinger-2022-large,
title = "What do Large Language Models Learn about Scripts?",
author = "Sancheti, Abhilasha and
Rudinger, Rachel",
editor = "Nastase, Vivi and
Pavlick, Ellie and
Pilehvar, Mohammad Taher and
Camacho-Collados, Jose and
Raganato, Alessandro",
booktitle = "Proceedings of the 11th Joint Conference on Lexical and Computational Semantics",
month = jul,
year = "2022",
address = "Seattle, Washington",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.starsem-1.1",
doi = "10.18653/v1/2022.starsem-1.1",
pages = "1--11",
abstract = "Script Knowledge (Schank and Abelson, 1975) has long been recognized as crucial for language understanding as it can help in filling in unstated information in a narrative. However, such knowledge is expensive to produce manually and difficult to induce from text due to reporting bias (Gordon and Van Durme, 2013). In this work, we are interested in the scientific question of whether explicit script knowledge is present and accessible through pre-trained generative language models (LMs). To this end, we introduce the task of generating full event sequence descriptions (ESDs) given a scenario as a natural language prompt. Through zero-shot probing, we find that generative LMs produce poor ESDs with mostly omitted, irrelevant, repeated or misordered events. To address this, we propose a pipeline-based script induction framework (SIF) which can generate good quality ESDs for unseen scenarios (e.g., bake a cake). SIF is a two-staged framework that fine-tunes LM on a small set of ESD examples in the first stage. In the second stage, ESD generated for an unseen scenario is post-processed using RoBERTa-based models to filter irrelevant events, remove repetitions, and reorder the temporally misordered events. Through automatic and manual evaluations, we demonstrate that SIF yields substantial improvements (1-3 BLEU points) over a fine-tuned LM. However, manual analysis shows that there is great room for improvement, offering a new research direction for inducing script knowledge.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sancheti-rudinger-2022-large">
<titleInfo>
<title>What do Large Language Models Learn about Scripts?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Abhilasha</namePart>
<namePart type="family">Sancheti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rachel</namePart>
<namePart type="family">Rudinger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 11th Joint Conference on Lexical and Computational Semantics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vivi</namePart>
<namePart type="family">Nastase</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ellie</namePart>
<namePart type="family">Pavlick</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jose</namePart>
<namePart type="family">Camacho-Collados</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Raganato</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Seattle, Washington</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Script Knowledge (Schank and Abelson, 1975) has long been recognized as crucial for language understanding as it can help in filling in unstated information in a narrative. However, such knowledge is expensive to produce manually and difficult to induce from text due to reporting bias (Gordon and Van Durme, 2013). In this work, we are interested in the scientific question of whether explicit script knowledge is present and accessible through pre-trained generative language models (LMs). To this end, we introduce the task of generating full event sequence descriptions (ESDs) given a scenario as a natural language prompt. Through zero-shot probing, we find that generative LMs produce poor ESDs with mostly omitted, irrelevant, repeated or misordered events. To address this, we propose a pipeline-based script induction framework (SIF) which can generate good quality ESDs for unseen scenarios (e.g., bake a cake). SIF is a two-staged framework that fine-tunes LM on a small set of ESD examples in the first stage. In the second stage, ESD generated for an unseen scenario is post-processed using RoBERTa-based models to filter irrelevant events, remove repetitions, and reorder the temporally misordered events. Through automatic and manual evaluations, we demonstrate that SIF yields substantial improvements (1-3 BLEU points) over a fine-tuned LM. However, manual analysis shows that there is great room for improvement, offering a new research direction for inducing script knowledge.</abstract>
<identifier type="citekey">sancheti-rudinger-2022-large</identifier>
<identifier type="doi">10.18653/v1/2022.starsem-1.1</identifier>
<location>
<url>https://aclanthology.org/2022.starsem-1.1</url>
</location>
<part>
<date>2022-07</date>
<extent unit="page">
<start>1</start>
<end>11</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T What do Large Language Models Learn about Scripts?
%A Sancheti, Abhilasha
%A Rudinger, Rachel
%Y Nastase, Vivi
%Y Pavlick, Ellie
%Y Pilehvar, Mohammad Taher
%Y Camacho-Collados, Jose
%Y Raganato, Alessandro
%S Proceedings of the 11th Joint Conference on Lexical and Computational Semantics
%D 2022
%8 July
%I Association for Computational Linguistics
%C Seattle, Washington
%F sancheti-rudinger-2022-large
%X Script Knowledge (Schank and Abelson, 1975) has long been recognized as crucial for language understanding as it can help in filling in unstated information in a narrative. However, such knowledge is expensive to produce manually and difficult to induce from text due to reporting bias (Gordon and Van Durme, 2013). In this work, we are interested in the scientific question of whether explicit script knowledge is present and accessible through pre-trained generative language models (LMs). To this end, we introduce the task of generating full event sequence descriptions (ESDs) given a scenario as a natural language prompt. Through zero-shot probing, we find that generative LMs produce poor ESDs with mostly omitted, irrelevant, repeated or misordered events. To address this, we propose a pipeline-based script induction framework (SIF) which can generate good quality ESDs for unseen scenarios (e.g., bake a cake). SIF is a two-staged framework that fine-tunes LM on a small set of ESD examples in the first stage. In the second stage, ESD generated for an unseen scenario is post-processed using RoBERTa-based models to filter irrelevant events, remove repetitions, and reorder the temporally misordered events. Through automatic and manual evaluations, we demonstrate that SIF yields substantial improvements (1-3 BLEU points) over a fine-tuned LM. However, manual analysis shows that there is great room for improvement, offering a new research direction for inducing script knowledge.
%R 10.18653/v1/2022.starsem-1.1
%U https://aclanthology.org/2022.starsem-1.1
%U https://doi.org/10.18653/v1/2022.starsem-1.1
%P 1-11
Markdown (Informal)
[What do Large Language Models Learn about Scripts?](https://aclanthology.org/2022.starsem-1.1) (Sancheti & Rudinger, *SEM 2022)
ACL
- Abhilasha Sancheti and Rachel Rudinger. 2022. What do Large Language Models Learn about Scripts?. In Proceedings of the 11th Joint Conference on Lexical and Computational Semantics, pages 1–11, Seattle, Washington. Association for Computational Linguistics.