<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="saparov-mitchell-2022-towards">
    <titleInfo>
        <title>Towards General Natural Language Understanding with Probabilistic Worldbuilding</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Abulhair</namePart>
        <namePart type="family">Saparov</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tom</namePart>
        <namePart type="given">M</namePart>
        <namePart type="family">Mitchell</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2022</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <genre authority="bibutilsgt">journal article</genre>
    <relatedItem type="host">
        <titleInfo>
            <title>Transactions of the Association for Computational Linguistics</title>
        </titleInfo>
        <originInfo>
            <issuance>continuing</issuance>
            <publisher>MIT Press</publisher>
            <place>
                <placeTerm type="text">Cambridge, MA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">periodical</genre>
        <genre authority="bibutilsgt">academic journal</genre>
    </relatedItem>
    <abstract>We introduce the Probabilistic Worldbuilding Model (PWM), a new fully symbolic Bayesian model of semantic parsing and reasoning, as a first step in a research program toward more domain- and task-general NLU and AI. Humans create internal mental models of their observations that greatly aid in their ability to understand and reason about a large variety of problems. In PWM, the meanings of sentences, acquired facts about the world, and intermediate steps in reasoning are all expressed in a human-readable formal language, with the design goal of interpretability. PWM is Bayesian, designed specifically to be able to generalize to new domains and new tasks. We derive and implement an inference algorithm that reads sentences by parsing and abducing updates to its latent world model that capture the semantics of those sentences, and evaluate it on two out-of-domain question-answering datasets: (1) ProofWriter and (2) a new dataset we call FictionalGeoQA, designed to be more representative of real language but still simple enough to focus on evaluating reasoning ability, while being robust against heuristics. Our method outperforms baselines on both, thereby demonstrating its value as a proof-of-concept.</abstract>
    <identifier type="citekey">saparov-mitchell-2022-towards</identifier>
    <identifier type="doi">10.1162/tacl_a_00463</identifier>
    <location>
        <url>https://aclanthology.org/2022.tacl-1.19</url>
    </location>
    <part>
        <date>2022</date>
        <detail type="volume"><number>10</number></detail>
        <extent unit="page">
            <start>325</start>
            <end>342</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Journal Article
%T Towards General Natural Language Understanding with Probabilistic Worldbuilding
%A Saparov, Abulhair
%A Mitchell, Tom M.
%J Transactions of the Association for Computational Linguistics
%D 2022
%V 10
%I MIT Press
%C Cambridge, MA
%F saparov-mitchell-2022-towards
%X We introduce the Probabilistic Worldbuilding Model (PWM), a new fully symbolic Bayesian model of semantic parsing and reasoning, as a first step in a research program toward more domain- and task-general NLU and AI. Humans create internal mental models of their observations that greatly aid in their ability to understand and reason about a large variety of problems. In PWM, the meanings of sentences, acquired facts about the world, and intermediate steps in reasoning are all expressed in a human-readable formal language, with the design goal of interpretability. PWM is Bayesian, designed specifically to be able to generalize to new domains and new tasks. We derive and implement an inference algorithm that reads sentences by parsing and abducing updates to its latent world model that capture the semantics of those sentences, and evaluate it on two out-of-domain question-answering datasets: (1) ProofWriter and (2) a new dataset we call FictionalGeoQA, designed to be more representative of real language but still simple enough to focus on evaluating reasoning ability, while being robust against heuristics. Our method outperforms baselines on both, thereby demonstrating its value as a proof-of-concept.
%R 10.1162/tacl_a_00463
%U https://aclanthology.org/2022.tacl-1.19
%U https://doi.org/10.1162/tacl_a_00463
%P 325-342

Markdown (Informal)

[Towards General Natural Language Understanding with Probabilistic Worldbuilding](https://aclanthology.org/2022.tacl-1.19) (Saparov & Mitchell, TACL 2022)

• Towards General Natural Language Understanding with Probabilistic Worldbuilding (Saparov & Mitchell, TACL 2022)

ACL

• Abulhair Saparov and Tom M. Mitchell. 2022. Towards General Natural Language Understanding with Probabilistic Worldbuilding. Transactions of the Association for Computational Linguistics, 10:325–342.