@inproceedings{tao-kamata-2026-teaching,
title = "Teaching Coding Assistants What to Change: Fine-Tuning from Pairwise Preferences",
author = "Tao, Hongmo and
Kamata, Hiroyuki",
editor = "Kamalloo, Ehsan and
Gontier, Nicolas and
Rozanov, Nikolai and
Hashemzadeh, Maryam and
Arora, Akhil",
booktitle = "Proceedings of the Second Workshop for {RE}search on Agent Language Models ({REALM} 2026)",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.realm-1.95/",
pages = "1363--1374",
ISBN = "979-8-89176-485-9",
abstract = "Real code-editing requests typically mix the target program with other related materials, such as test results or documentation. However, models often make mistakes by incorrectly changing these supporting materials or missing the actual code that needs correction. To investigate this issue, we created multiple task scenarios where the code needing changes and the intended behavior remained constant, while only the surrounding supporting texts varied. This approach allowed us to test if supporting texts that resemble code can confuse the model, making it difficult to identify the correct code to modify. To address this, we developed a simple yet effective training method by showing the model examples of correct and incorrect edits. Specifically, Correct examples only modify the intended code, whereas incorrect examples mistakenly alter or duplicate supporting texts. Our results suggest that reliable code editing depends on the model{'}s early ability to distinguish between guidance materials and the actual code requiring changes."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tao-kamata-2026-teaching">
<titleInfo>
<title>Teaching Coding Assistants What to Change: Fine-Tuning from Pairwise Preferences</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hongmo</namePart>
<namePart type="family">Tao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hiroyuki</namePart>
<namePart type="family">Kamata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Second Workshop for REsearch on Agent Language Models (REALM 2026)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ehsan</namePart>
<namePart type="family">Kamalloo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicolas</namePart>
<namePart type="family">Gontier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolai</namePart>
<namePart type="family">Rozanov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maryam</namePart>
<namePart type="family">Hashemzadeh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Akhil</namePart>
<namePart type="family">Arora</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-485-9</identifier>
</relatedItem>
<abstract>Real code-editing requests typically mix the target program with other related materials, such as test results or documentation. However, models often make mistakes by incorrectly changing these supporting materials or missing the actual code that needs correction. To investigate this issue, we created multiple task scenarios where the code needing changes and the intended behavior remained constant, while only the surrounding supporting texts varied. This approach allowed us to test if supporting texts that resemble code can confuse the model, making it difficult to identify the correct code to modify. To address this, we developed a simple yet effective training method by showing the model examples of correct and incorrect edits. Specifically, Correct examples only modify the intended code, whereas incorrect examples mistakenly alter or duplicate supporting texts. Our results suggest that reliable code editing depends on the model’s early ability to distinguish between guidance materials and the actual code requiring changes.</abstract>
<identifier type="citekey">tao-kamata-2026-teaching</identifier>
<location>
<url>https://aclanthology.org/2026.realm-1.95/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>1363</start>
<end>1374</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Teaching Coding Assistants What to Change: Fine-Tuning from Pairwise Preferences
%A Tao, Hongmo
%A Kamata, Hiroyuki
%Y Kamalloo, Ehsan
%Y Gontier, Nicolas
%Y Rozanov, Nikolai
%Y Hashemzadeh, Maryam
%Y Arora, Akhil
%S Proceedings of the Second Workshop for REsearch on Agent Language Models (REALM 2026)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-485-9
%F tao-kamata-2026-teaching
%X Real code-editing requests typically mix the target program with other related materials, such as test results or documentation. However, models often make mistakes by incorrectly changing these supporting materials or missing the actual code that needs correction. To investigate this issue, we created multiple task scenarios where the code needing changes and the intended behavior remained constant, while only the surrounding supporting texts varied. This approach allowed us to test if supporting texts that resemble code can confuse the model, making it difficult to identify the correct code to modify. To address this, we developed a simple yet effective training method by showing the model examples of correct and incorrect edits. Specifically, Correct examples only modify the intended code, whereas incorrect examples mistakenly alter or duplicate supporting texts. Our results suggest that reliable code editing depends on the model’s early ability to distinguish between guidance materials and the actual code requiring changes.
%U https://aclanthology.org/2026.realm-1.95/
%P 1363-1374
Markdown (Informal)
[Teaching Coding Assistants What to Change: Fine-Tuning from Pairwise Preferences](https://aclanthology.org/2026.realm-1.95/) (Tao & Kamata, REALM 2026)
ACL