@inproceedings{sanyal-etal-2024-machines,
title = "Are Machines Better at Complex Reasoning? Unveiling Human-Machine Inference Gaps in Entailment Verification",
author = "Sanyal, Soumya and
Xiao, Tianyi and
Liu, Jiacheng and
Wang, Wenya and
Ren, Xiang",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand and virtual meeting",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.618",
pages = "10361--10386",
abstract = "Making inferences in text comprehension to understand the meaning is essential in language processing. This work studies the entailment verification (EV) problem of complex, multi-sentence premises requiring a system to make multiple inferences implicitly. Modern applications of EV in detecting inconsistent model-generated rationales require complex multi-hop reasoning. However, current textual inference datasets mostly contain short-sentence premises that partially focus on this. To address this, we compile an EV benchmark that includes datasets from three NLP domains (NLI, contextual QA, and rationales) containing multi-sentence premises. On benchmarking humans and LLMs, we find that LLMs are better than humans in multi-hop reasoning across extended contexts, while humans perform better in simple deductive reasoning tasks. We also finetune a Flan-T5 model for EV using two training objectives to obtain a strong open-source model that outperforms GPT-3.5 and rivals GPT-4. Finally, we use our finetuned model to filter out inconsistent model-generated rationales in self-consistency decoding, resulting in a 6{\%} accuracy improvement on average across three MCQ datasets.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sanyal-etal-2024-machines">
<titleInfo>
<title>Are Machines Better at Complex Reasoning? Unveiling Human-Machine Inference Gaps in Entailment Verification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Soumya</namePart>
<namePart type="family">Sanyal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tianyi</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiacheng</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenya</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiang</namePart>
<namePart type="family">Ren</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics ACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand and virtual meeting</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Making inferences in text comprehension to understand the meaning is essential in language processing. This work studies the entailment verification (EV) problem of complex, multi-sentence premises requiring a system to make multiple inferences implicitly. Modern applications of EV in detecting inconsistent model-generated rationales require complex multi-hop reasoning. However, current textual inference datasets mostly contain short-sentence premises that partially focus on this. To address this, we compile an EV benchmark that includes datasets from three NLP domains (NLI, contextual QA, and rationales) containing multi-sentence premises. On benchmarking humans and LLMs, we find that LLMs are better than humans in multi-hop reasoning across extended contexts, while humans perform better in simple deductive reasoning tasks. We also finetune a Flan-T5 model for EV using two training objectives to obtain a strong open-source model that outperforms GPT-3.5 and rivals GPT-4. Finally, we use our finetuned model to filter out inconsistent model-generated rationales in self-consistency decoding, resulting in a 6% accuracy improvement on average across three MCQ datasets.</abstract>
<identifier type="citekey">sanyal-etal-2024-machines</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.618</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>10361</start>
<end>10386</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Are Machines Better at Complex Reasoning? Unveiling Human-Machine Inference Gaps in Entailment Verification
%A Sanyal, Soumya
%A Xiao, Tianyi
%A Liu, Jiacheng
%A Wang, Wenya
%A Ren, Xiang
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand and virtual meeting
%F sanyal-etal-2024-machines
%X Making inferences in text comprehension to understand the meaning is essential in language processing. This work studies the entailment verification (EV) problem of complex, multi-sentence premises requiring a system to make multiple inferences implicitly. Modern applications of EV in detecting inconsistent model-generated rationales require complex multi-hop reasoning. However, current textual inference datasets mostly contain short-sentence premises that partially focus on this. To address this, we compile an EV benchmark that includes datasets from three NLP domains (NLI, contextual QA, and rationales) containing multi-sentence premises. On benchmarking humans and LLMs, we find that LLMs are better than humans in multi-hop reasoning across extended contexts, while humans perform better in simple deductive reasoning tasks. We also finetune a Flan-T5 model for EV using two training objectives to obtain a strong open-source model that outperforms GPT-3.5 and rivals GPT-4. Finally, we use our finetuned model to filter out inconsistent model-generated rationales in self-consistency decoding, resulting in a 6% accuracy improvement on average across three MCQ datasets.
%U https://aclanthology.org/2024.findings-acl.618
%P 10361-10386
Markdown (Informal)
[Are Machines Better at Complex Reasoning? Unveiling Human-Machine Inference Gaps in Entailment Verification](https://aclanthology.org/2024.findings-acl.618) (Sanyal et al., Findings 2024)
ACL