@article{lehmann-etal-2024-beyond,
title = "Beyond Boundaries: A Human-like Approach for Question Answering over Structured and Unstructured Information Sources",
author = "Lehmann, Jens and
Bhandiwad, Dhananjay and
Gattogi, Preetam and
Vahdati, Sahar",
journal = "Transactions of the Association for Computational Linguistics",
volume = "12",
year = "2024",
address = "Cambridge, MA",
publisher = "MIT Press",
url = "https://aclanthology.org/2024.tacl-1.44",
doi = "10.1162/tacl_a_00671",
pages = "786--802",
abstract = "Answering factual questions from heterogenous sources, such as graphs and text, is a key capacity of intelligent systems. Current approaches either (i) perform question answering over text and structured sources as separate pipelines followed by a merge step or (ii) provide an early integration, giving up the strengths of particular information sources. To solve this problem, we present {``}HumanIQ{''}, a method that teaches language models to dynamically combine retrieved information by imitating how humans use retrieval tools. Our approach couples a generic method for gathering human demonstrations of tool use with adaptive few-shot learning for tool augmented models. We show that HumanIQ confers significant benefits, including i) reducing the error rate of our strongest baseline (GPT-4) by over 50{\%} across 3 benchmarks, (ii) improving human preference over responses from vanilla GPT-4 (45.3{\%} wins, 46.7{\%} ties, 8.0{\%} loss), and (iii) outperforming numerous task-specific baselines.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lehmann-etal-2024-beyond">
<titleInfo>
<title>Beyond Boundaries: A Human-like Approach for Question Answering over Structured and Unstructured Information Sources</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jens</namePart>
<namePart type="family">Lehmann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dhananjay</namePart>
<namePart type="family">Bhandiwad</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Preetam</namePart>
<namePart type="family">Gattogi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sahar</namePart>
<namePart type="family">Vahdati</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024</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>Answering factual questions from heterogenous sources, such as graphs and text, is a key capacity of intelligent systems. Current approaches either (i) perform question answering over text and structured sources as separate pipelines followed by a merge step or (ii) provide an early integration, giving up the strengths of particular information sources. To solve this problem, we present “HumanIQ”, a method that teaches language models to dynamically combine retrieved information by imitating how humans use retrieval tools. Our approach couples a generic method for gathering human demonstrations of tool use with adaptive few-shot learning for tool augmented models. We show that HumanIQ confers significant benefits, including i) reducing the error rate of our strongest baseline (GPT-4) by over 50% across 3 benchmarks, (ii) improving human preference over responses from vanilla GPT-4 (45.3% wins, 46.7% ties, 8.0% loss), and (iii) outperforming numerous task-specific baselines.</abstract>
<identifier type="citekey">lehmann-etal-2024-beyond</identifier>
<identifier type="doi">10.1162/tacl_a_00671</identifier>
<location>
<url>https://aclanthology.org/2024.tacl-1.44</url>
</location>
<part>
<date>2024</date>
<detail type="volume"><number>12</number></detail>
<extent unit="page">
<start>786</start>
<end>802</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Journal Article
%T Beyond Boundaries: A Human-like Approach for Question Answering over Structured and Unstructured Information Sources
%A Lehmann, Jens
%A Bhandiwad, Dhananjay
%A Gattogi, Preetam
%A Vahdati, Sahar
%J Transactions of the Association for Computational Linguistics
%D 2024
%V 12
%I MIT Press
%C Cambridge, MA
%F lehmann-etal-2024-beyond
%X Answering factual questions from heterogenous sources, such as graphs and text, is a key capacity of intelligent systems. Current approaches either (i) perform question answering over text and structured sources as separate pipelines followed by a merge step or (ii) provide an early integration, giving up the strengths of particular information sources. To solve this problem, we present “HumanIQ”, a method that teaches language models to dynamically combine retrieved information by imitating how humans use retrieval tools. Our approach couples a generic method for gathering human demonstrations of tool use with adaptive few-shot learning for tool augmented models. We show that HumanIQ confers significant benefits, including i) reducing the error rate of our strongest baseline (GPT-4) by over 50% across 3 benchmarks, (ii) improving human preference over responses from vanilla GPT-4 (45.3% wins, 46.7% ties, 8.0% loss), and (iii) outperforming numerous task-specific baselines.
%R 10.1162/tacl_a_00671
%U https://aclanthology.org/2024.tacl-1.44
%U https://doi.org/10.1162/tacl_a_00671
%P 786-802
Markdown (Informal)
[Beyond Boundaries: A Human-like Approach for Question Answering over Structured and Unstructured Information Sources](https://aclanthology.org/2024.tacl-1.44) (Lehmann et al., TACL 2024)
ACL