@inproceedings{jobanputra-etal-2025-proofteller,
title = "{P}roof{T}eller: Exposing recency bias in {LLM} reasoning and its side effects on communication",
author = "Jobanputra, Mayank and
Kovtunova, Alisa and
Balthes, Brisca and
Pogulskiy, Fedor Grigoryevich and
Wang, Yifan and
Borgwardt, Stefan and
Demberg, Vera",
editor = "Inui, Kentaro and
Sakti, Sakriani and
Wang, Haofen and
Wong, Derek F. and
Bhattacharyya, Pushpak and
Banerjee, Biplab and
Ekbal, Asif and
Chakraborty, Tanmoy and
Singh, Dhirendra Pratap",
booktitle = "Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics",
month = dec,
year = "2025",
address = "Mumbai, India",
publisher = "The Asian Federation of Natural Language Processing and The Association for Computational Linguistics",
url = "https://aclanthology.org/2025.ijcnlp-long.80/",
pages = "1439--1462",
ISBN = "979-8-89176-298-5",
abstract = "Large language models (LLMs) are increasingly applied in domains that demand reliable and interpretable reasoning. While formal methods can generate provably correct proofs, these proofs are often inaccessible to non-expert users. This raises a natural question: can LLMs, when given a verified proof, faithfully interpret its reasoning and communicate it clearly? We introduce $ProofTeller$, a benchmark that evaluates this ability across three tasks: (1) identifying key proof steps, (2) summarizing the reasoning, and (3) explaining the result in concise natural language. The benchmark covers three domains: {\_}Biology{\_}, {\_}Drones{\_}, and {\_}Recipes{\_}, representing scientific, safety-critical, and everyday reasoning scenarios. We find a consistent near-conclusion bias: LLMs tend to focus on steps closest to the final proof conclusion rather than on the most informative ones. A targeted human study confirms that explanations based on such steps are rated less appropriate for end users. These findings indicate that even when reasoning is provided, current LLMs face challenges in communicating key information in a useful manner, highlighting the need for LLMs that can communicate important details reliably."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jobanputra-etal-2025-proofteller">
<titleInfo>
<title>ProofTeller: Exposing recency bias in LLM reasoning and its side effects on communication</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mayank</namePart>
<namePart type="family">Jobanputra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alisa</namePart>
<namePart type="family">Kovtunova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Brisca</namePart>
<namePart type="family">Balthes</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fedor</namePart>
<namePart type="given">Grigoryevich</namePart>
<namePart type="family">Pogulskiy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yifan</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stefan</namePart>
<namePart type="family">Borgwardt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vera</namePart>
<namePart type="family">Demberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haofen</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Derek</namePart>
<namePart type="given">F</namePart>
<namePart type="family">Wong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pushpak</namePart>
<namePart type="family">Bhattacharyya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Biplab</namePart>
<namePart type="family">Banerjee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Asif</namePart>
<namePart type="family">Ekbal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dhirendra</namePart>
<namePart type="given">Pratap</namePart>
<namePart type="family">Singh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>The Asian Federation of Natural Language Processing and The Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Mumbai, India</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-298-5</identifier>
</relatedItem>
<abstract>Large language models (LLMs) are increasingly applied in domains that demand reliable and interpretable reasoning. While formal methods can generate provably correct proofs, these proofs are often inaccessible to non-expert users. This raises a natural question: can LLMs, when given a verified proof, faithfully interpret its reasoning and communicate it clearly? We introduce ProofTeller, a benchmark that evaluates this ability across three tasks: (1) identifying key proof steps, (2) summarizing the reasoning, and (3) explaining the result in concise natural language. The benchmark covers three domains: _Biology_, _Drones_, and _Recipes_, representing scientific, safety-critical, and everyday reasoning scenarios. We find a consistent near-conclusion bias: LLMs tend to focus on steps closest to the final proof conclusion rather than on the most informative ones. A targeted human study confirms that explanations based on such steps are rated less appropriate for end users. These findings indicate that even when reasoning is provided, current LLMs face challenges in communicating key information in a useful manner, highlighting the need for LLMs that can communicate important details reliably.</abstract>
<identifier type="citekey">jobanputra-etal-2025-proofteller</identifier>
<location>
<url>https://aclanthology.org/2025.ijcnlp-long.80/</url>
</location>
<part>
<date>2025-12</date>
<extent unit="page">
<start>1439</start>
<end>1462</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ProofTeller: Exposing recency bias in LLM reasoning and its side effects on communication
%A Jobanputra, Mayank
%A Kovtunova, Alisa
%A Balthes, Brisca
%A Pogulskiy, Fedor Grigoryevich
%A Wang, Yifan
%A Borgwardt, Stefan
%A Demberg, Vera
%Y Inui, Kentaro
%Y Sakti, Sakriani
%Y Wang, Haofen
%Y Wong, Derek F.
%Y Bhattacharyya, Pushpak
%Y Banerjee, Biplab
%Y Ekbal, Asif
%Y Chakraborty, Tanmoy
%Y Singh, Dhirendra Pratap
%S Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics
%D 2025
%8 December
%I The Asian Federation of Natural Language Processing and The Association for Computational Linguistics
%C Mumbai, India
%@ 979-8-89176-298-5
%F jobanputra-etal-2025-proofteller
%X Large language models (LLMs) are increasingly applied in domains that demand reliable and interpretable reasoning. While formal methods can generate provably correct proofs, these proofs are often inaccessible to non-expert users. This raises a natural question: can LLMs, when given a verified proof, faithfully interpret its reasoning and communicate it clearly? We introduce ProofTeller, a benchmark that evaluates this ability across three tasks: (1) identifying key proof steps, (2) summarizing the reasoning, and (3) explaining the result in concise natural language. The benchmark covers three domains: _Biology_, _Drones_, and _Recipes_, representing scientific, safety-critical, and everyday reasoning scenarios. We find a consistent near-conclusion bias: LLMs tend to focus on steps closest to the final proof conclusion rather than on the most informative ones. A targeted human study confirms that explanations based on such steps are rated less appropriate for end users. These findings indicate that even when reasoning is provided, current LLMs face challenges in communicating key information in a useful manner, highlighting the need for LLMs that can communicate important details reliably.
%U https://aclanthology.org/2025.ijcnlp-long.80/
%P 1439-1462
Markdown (Informal)
[ProofTeller: Exposing recency bias in LLM reasoning and its side effects on communication](https://aclanthology.org/2025.ijcnlp-long.80/) (Jobanputra et al., IJCNLP-AACL 2025)
ACL
- Mayank Jobanputra, Alisa Kovtunova, Brisca Balthes, Fedor Grigoryevich Pogulskiy, Yifan Wang, Stefan Borgwardt, and Vera Demberg. 2025. ProofTeller: Exposing recency bias in LLM reasoning and its side effects on communication. In Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics, pages 1439–1462, Mumbai, India. The Asian Federation of Natural Language Processing and The Association for Computational Linguistics.