@inproceedings{dalal-etal-2024-inference,
title = "Inference to the Best Explanation in Large Language Models",
author = "Dalal, Dhairya and
Valentino, Marco and
Freitas, Andre and
Buitelaar, Paul",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.acl-long.14",
doi = "10.18653/v1/2024.acl-long.14",
pages = "217--235",
abstract = "While Large Language Models (LLMs) have found success in real-world applications, their underlying explanatory process is still poorly understood. This paper proposes \textit{IBE-Eval}, a framework inspired by philosophical accounts on \textit{Inference to the Best Explanation (IBE)} to advance the interpretation and evaluation of LLMs{'} explanations. \textit{IBE-Eval} estimates the plausibility of natural language explanations through a combination of explicit logical and linguistic features including: \textit{consistency}, \textit{parsimony}, \textit{coherence}, and \textit{uncertainty}. Extensive experiments are conducted on \textit{Causal Question Answering (CQA)}, where \textit{IBE-Eval} is tasked to select the most plausible causal explanation amongst competing ones generated by LLMs (i.e., GPT 3.5 and Llama 2). The experiments reveal that \textit{IBE-Eval} can successfully identify the best explanation with up to 77{\%} accuracy ($\approx 27\%$ above random), improving upon a GPT 3.5-as-a-Judge baseline ($\approx+17\%$) while being intrinsically more efficient and interpretable. Additional analyses suggest that, despite model-specific variances, LLM-generated explanations tend to conform to IBE criteria and that \textit{IBE-Eval} is significantly correlated with human judgment, opening up opportunities for future development of automated explanation verification tools.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dalal-etal-2024-inference">
<titleInfo>
<title>Inference to the Best Explanation in Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dhairya</namePart>
<namePart type="family">Dalal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Valentino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Freitas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="family">Buitelaar</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>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</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</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>While Large Language Models (LLMs) have found success in real-world applications, their underlying explanatory process is still poorly understood. This paper proposes IBE-Eval, a framework inspired by philosophical accounts on Inference to the Best Explanation (IBE) to advance the interpretation and evaluation of LLMs’ explanations. IBE-Eval estimates the plausibility of natural language explanations through a combination of explicit logical and linguistic features including: consistency, parsimony, coherence, and uncertainty. Extensive experiments are conducted on Causal Question Answering (CQA), where IBE-Eval is tasked to select the most plausible causal explanation amongst competing ones generated by LLMs (i.e., GPT 3.5 and Llama 2). The experiments reveal that IBE-Eval can successfully identify the best explanation with up to 77% accuracy (\approx 27% above random), improving upon a GPT 3.5-as-a-Judge baseline (\approx+17%) while being intrinsically more efficient and interpretable. Additional analyses suggest that, despite model-specific variances, LLM-generated explanations tend to conform to IBE criteria and that IBE-Eval is significantly correlated with human judgment, opening up opportunities for future development of automated explanation verification tools.</abstract>
<identifier type="citekey">dalal-etal-2024-inference</identifier>
<identifier type="doi">10.18653/v1/2024.acl-long.14</identifier>
<location>
<url>https://aclanthology.org/2024.acl-long.14</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>217</start>
<end>235</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Inference to the Best Explanation in Large Language Models
%A Dalal, Dhairya
%A Valentino, Marco
%A Freitas, Andre
%A Buitelaar, Paul
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F dalal-etal-2024-inference
%X While Large Language Models (LLMs) have found success in real-world applications, their underlying explanatory process is still poorly understood. This paper proposes IBE-Eval, a framework inspired by philosophical accounts on Inference to the Best Explanation (IBE) to advance the interpretation and evaluation of LLMs’ explanations. IBE-Eval estimates the plausibility of natural language explanations through a combination of explicit logical and linguistic features including: consistency, parsimony, coherence, and uncertainty. Extensive experiments are conducted on Causal Question Answering (CQA), where IBE-Eval is tasked to select the most plausible causal explanation amongst competing ones generated by LLMs (i.e., GPT 3.5 and Llama 2). The experiments reveal that IBE-Eval can successfully identify the best explanation with up to 77% accuracy (\approx 27% above random), improving upon a GPT 3.5-as-a-Judge baseline (\approx+17%) while being intrinsically more efficient and interpretable. Additional analyses suggest that, despite model-specific variances, LLM-generated explanations tend to conform to IBE criteria and that IBE-Eval is significantly correlated with human judgment, opening up opportunities for future development of automated explanation verification tools.
%R 10.18653/v1/2024.acl-long.14
%U https://aclanthology.org/2024.acl-long.14
%U https://doi.org/10.18653/v1/2024.acl-long.14
%P 217-235
Markdown (Informal)
[Inference to the Best Explanation in Large Language Models](https://aclanthology.org/2024.acl-long.14) (Dalal et al., ACL 2024)
ACL
- Dhairya Dalal, Marco Valentino, Andre Freitas, and Paul Buitelaar. 2024. Inference to the Best Explanation in Large Language Models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 217–235, Bangkok, Thailand. Association for Computational Linguistics.