@inproceedings{liang-surdeanu-2020-transformers,
title = "Do Transformers Dream of Inference, or Can Pretrained Generative Models Learn Implicit Inferential Rules?",
author = "Liang, Zhengzhong and
Surdeanu, Mihai",
editor = "Rogers, Anna and
Sedoc, Jo{\~a}o and
Rumshisky, Anna",
booktitle = "Proceedings of the First Workshop on Insights from Negative Results in NLP",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.insights-1.12",
doi = "10.18653/v1/2020.insights-1.12",
pages = "76--81",
abstract = "Large pretrained language models (LM) have been used successfully for multi-hop question answering. However, most of these directions are not interpretable, as they do not make the inference hops necessary to explain a candidate answer explicitly. In this work, we investigate the capability of a state-of-the-art transformer LM to generate explicit inference hops, i.e., to infer a new statement necessary to answer a question given some premise input statements. Our analysis shows that such LMs can generate new statements for some simple inference types, but performance remains poor for complex, real-world inference types such as those that require monotonicity, composition, and commonsense knowledge.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liang-surdeanu-2020-transformers">
<titleInfo>
<title>Do Transformers Dream of Inference, or Can Pretrained Generative Models Learn Implicit Inferential Rules?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhengzhong</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mihai</namePart>
<namePart type="family">Surdeanu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the First Workshop on Insights from Negative Results in NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">João</namePart>
<namePart type="family">Sedoc</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rumshisky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large pretrained language models (LM) have been used successfully for multi-hop question answering. However, most of these directions are not interpretable, as they do not make the inference hops necessary to explain a candidate answer explicitly. In this work, we investigate the capability of a state-of-the-art transformer LM to generate explicit inference hops, i.e., to infer a new statement necessary to answer a question given some premise input statements. Our analysis shows that such LMs can generate new statements for some simple inference types, but performance remains poor for complex, real-world inference types such as those that require monotonicity, composition, and commonsense knowledge.</abstract>
<identifier type="citekey">liang-surdeanu-2020-transformers</identifier>
<identifier type="doi">10.18653/v1/2020.insights-1.12</identifier>
<location>
<url>https://aclanthology.org/2020.insights-1.12</url>
</location>
<part>
<date>2020-11</date>
<extent unit="page">
<start>76</start>
<end>81</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Do Transformers Dream of Inference, or Can Pretrained Generative Models Learn Implicit Inferential Rules?
%A Liang, Zhengzhong
%A Surdeanu, Mihai
%Y Rogers, Anna
%Y Sedoc, João
%Y Rumshisky, Anna
%S Proceedings of the First Workshop on Insights from Negative Results in NLP
%D 2020
%8 November
%I Association for Computational Linguistics
%C Online
%F liang-surdeanu-2020-transformers
%X Large pretrained language models (LM) have been used successfully for multi-hop question answering. However, most of these directions are not interpretable, as they do not make the inference hops necessary to explain a candidate answer explicitly. In this work, we investigate the capability of a state-of-the-art transformer LM to generate explicit inference hops, i.e., to infer a new statement necessary to answer a question given some premise input statements. Our analysis shows that such LMs can generate new statements for some simple inference types, but performance remains poor for complex, real-world inference types such as those that require monotonicity, composition, and commonsense knowledge.
%R 10.18653/v1/2020.insights-1.12
%U https://aclanthology.org/2020.insights-1.12
%U https://doi.org/10.18653/v1/2020.insights-1.12
%P 76-81
Markdown (Informal)
[Do Transformers Dream of Inference, or Can Pretrained Generative Models Learn Implicit Inferential Rules?](https://aclanthology.org/2020.insights-1.12) (Liang & Surdeanu, insights 2020)
ACL