@inproceedings{furman-etal-2022-sequence,
title = "A Sequence Modelling Approach to Question Answering in Text-Based Games",
author = "Furman, Gregory and
Toledo, Edan and
Shock, Jonathan and
Buys, Jan",
editor = "C{\^o}t{\'e}, Marc-Alexandre and
Yuan, Xingdi and
Ammanabrolu, Prithviraj",
booktitle = "Proceedings of the 3rd Wordplay: When Language Meets Games Workshop (Wordplay 2022)",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.wordplay-1.4/",
doi = "10.18653/v1/2022.wordplay-1.4",
pages = "44--58",
abstract = "Interactive Question Answering (IQA) requires an intelligent agent to interact with a dynamic environment in order to gather information necessary to answer a question. IQA tasks have been proposed as means of training systems to develop language or visual comprehension abilities. To this end, the Question Answering with Interactive Text (QAit) task was created to produce and benchmark interactive agents capable of seeking information and answering questions in unseen environments. While prior work has exclusively focused on IQA as a reinforcement learning problem, such methods suffer from low sample efficiency and poor accuracy in zero-shot evaluation. In this paper, we propose the use of the recently proposed Decision Transformer architecture to provide improvements upon prior baselines. By utilising a causally masked GPT-2 Transformer for command generation and a BERT model for question answer prediction, we show that the Decision Transformer achieves performance greater than or equal to current state-of-the-art RL baselines on the QAit task in a sample efficient manner. In addition, these results are achievable by training on sub-optimal random trajectories, therefore not requiring the use of online agents to gather data."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="furman-etal-2022-sequence">
<titleInfo>
<title>A Sequence Modelling Approach to Question Answering in Text-Based Games</title>
</titleInfo>
<name type="personal">
<namePart type="given">Gregory</namePart>
<namePart type="family">Furman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Edan</namePart>
<namePart type="family">Toledo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jonathan</namePart>
<namePart type="family">Shock</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Buys</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 3rd Wordplay: When Language Meets Games Workshop (Wordplay 2022)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marc-Alexandre</namePart>
<namePart type="family">Côté</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xingdi</namePart>
<namePart type="family">Yuan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Prithviraj</namePart>
<namePart type="family">Ammanabrolu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Seattle, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Interactive Question Answering (IQA) requires an intelligent agent to interact with a dynamic environment in order to gather information necessary to answer a question. IQA tasks have been proposed as means of training systems to develop language or visual comprehension abilities. To this end, the Question Answering with Interactive Text (QAit) task was created to produce and benchmark interactive agents capable of seeking information and answering questions in unseen environments. While prior work has exclusively focused on IQA as a reinforcement learning problem, such methods suffer from low sample efficiency and poor accuracy in zero-shot evaluation. In this paper, we propose the use of the recently proposed Decision Transformer architecture to provide improvements upon prior baselines. By utilising a causally masked GPT-2 Transformer for command generation and a BERT model for question answer prediction, we show that the Decision Transformer achieves performance greater than or equal to current state-of-the-art RL baselines on the QAit task in a sample efficient manner. In addition, these results are achievable by training on sub-optimal random trajectories, therefore not requiring the use of online agents to gather data.</abstract>
<identifier type="citekey">furman-etal-2022-sequence</identifier>
<identifier type="doi">10.18653/v1/2022.wordplay-1.4</identifier>
<location>
<url>https://aclanthology.org/2022.wordplay-1.4/</url>
</location>
<part>
<date>2022-07</date>
<extent unit="page">
<start>44</start>
<end>58</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Sequence Modelling Approach to Question Answering in Text-Based Games
%A Furman, Gregory
%A Toledo, Edan
%A Shock, Jonathan
%A Buys, Jan
%Y Côté, Marc-Alexandre
%Y Yuan, Xingdi
%Y Ammanabrolu, Prithviraj
%S Proceedings of the 3rd Wordplay: When Language Meets Games Workshop (Wordplay 2022)
%D 2022
%8 July
%I Association for Computational Linguistics
%C Seattle, United States
%F furman-etal-2022-sequence
%X Interactive Question Answering (IQA) requires an intelligent agent to interact with a dynamic environment in order to gather information necessary to answer a question. IQA tasks have been proposed as means of training systems to develop language or visual comprehension abilities. To this end, the Question Answering with Interactive Text (QAit) task was created to produce and benchmark interactive agents capable of seeking information and answering questions in unseen environments. While prior work has exclusively focused on IQA as a reinforcement learning problem, such methods suffer from low sample efficiency and poor accuracy in zero-shot evaluation. In this paper, we propose the use of the recently proposed Decision Transformer architecture to provide improvements upon prior baselines. By utilising a causally masked GPT-2 Transformer for command generation and a BERT model for question answer prediction, we show that the Decision Transformer achieves performance greater than or equal to current state-of-the-art RL baselines on the QAit task in a sample efficient manner. In addition, these results are achievable by training on sub-optimal random trajectories, therefore not requiring the use of online agents to gather data.
%R 10.18653/v1/2022.wordplay-1.4
%U https://aclanthology.org/2022.wordplay-1.4/
%U https://doi.org/10.18653/v1/2022.wordplay-1.4
%P 44-58
Markdown (Informal)
[A Sequence Modelling Approach to Question Answering in Text-Based Games](https://aclanthology.org/2022.wordplay-1.4/) (Furman et al., Wordplay 2022)
ACL