@inproceedings{yongsatianchot-etal-2026-thaieditbench,
title = "{T}hai{E}dit{B}ench: A Mechanism-Typed, Length-Stratified Benchmark for {T}hai Orthographic Error Correction",
author = "Yongsatianchot, Nutchanon and
Ittichaiwong, Piyalitt and
Veerakanjana, Kanyakorn",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 2: Short Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-short.17/",
pages = "211--225",
ISBN = "979-8-89176-440-8",
abstract = "Thai is written without spaces between words, and its spelling depends on tone marks, silent consonants, and the transliteration of many loanwords. Each of these is a common source of misspellings. We present ThaiEditBench, a benchmark for correcting Thai spelling errors at the character level. It has three components: (i) a six-class taxonomy that types each error by the part of the spelling that goes wrong, plus an etymology tag; (ii) three length tiers (sentence, paragraph, and page) that test whether a model stays accurate over long, mostly correct text; and (iii) a contamination-robust construction, in which naturally attested misspellings are injected into clean text, so the corrupted passages do not appear verbatim in the source corpora. Because the errors are injected, the gold corrections are exact, and a blind two-annotator study measures how reliably the taxonomy can be applied. Grading is at the level of Thai character clusters, with span-based matching adapted from ERRANT. We evaluate 16 models and find that (1) under the benchmark{'}s conditions, a model that locates an error almost always fixes it correctly, so what separates models is detection; (2) the evaluated Thai-adapted models do not lead the strongest general models; and (3) in long, mostly correct passages, weaker models accumulate over-edits that pull their scores far below their sentence-level results. On real errors mined from Wikipedia edit history, scores are somewhat lower, but the model ranking largely holds."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yongsatianchot-etal-2026-thaieditbench">
<titleInfo>
<title>ThaiEditBench: A Mechanism-Typed, Length-Stratified Benchmark for Thai Orthographic Error Correction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nutchanon</namePart>
<namePart type="family">Yongsatianchot</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Piyalitt</namePart>
<namePart type="family">Ittichaiwong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kanyakorn</namePart>
<namePart type="family">Veerakanjana</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 2: Short Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-440-8</identifier>
</relatedItem>
<abstract>Thai is written without spaces between words, and its spelling depends on tone marks, silent consonants, and the transliteration of many loanwords. Each of these is a common source of misspellings. We present ThaiEditBench, a benchmark for correcting Thai spelling errors at the character level. It has three components: (i) a six-class taxonomy that types each error by the part of the spelling that goes wrong, plus an etymology tag; (ii) three length tiers (sentence, paragraph, and page) that test whether a model stays accurate over long, mostly correct text; and (iii) a contamination-robust construction, in which naturally attested misspellings are injected into clean text, so the corrupted passages do not appear verbatim in the source corpora. Because the errors are injected, the gold corrections are exact, and a blind two-annotator study measures how reliably the taxonomy can be applied. Grading is at the level of Thai character clusters, with span-based matching adapted from ERRANT. We evaluate 16 models and find that (1) under the benchmark’s conditions, a model that locates an error almost always fixes it correctly, so what separates models is detection; (2) the evaluated Thai-adapted models do not lead the strongest general models; and (3) in long, mostly correct passages, weaker models accumulate over-edits that pull their scores far below their sentence-level results. On real errors mined from Wikipedia edit history, scores are somewhat lower, but the model ranking largely holds.</abstract>
<identifier type="citekey">yongsatianchot-etal-2026-thaieditbench</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-short.17/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>211</start>
<end>225</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ThaiEditBench: A Mechanism-Typed, Length-Stratified Benchmark for Thai Orthographic Error Correction
%A Yongsatianchot, Nutchanon
%A Ittichaiwong, Piyalitt
%A Veerakanjana, Kanyakorn
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 2: Short Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-440-8
%F yongsatianchot-etal-2026-thaieditbench
%X Thai is written without spaces between words, and its spelling depends on tone marks, silent consonants, and the transliteration of many loanwords. Each of these is a common source of misspellings. We present ThaiEditBench, a benchmark for correcting Thai spelling errors at the character level. It has three components: (i) a six-class taxonomy that types each error by the part of the spelling that goes wrong, plus an etymology tag; (ii) three length tiers (sentence, paragraph, and page) that test whether a model stays accurate over long, mostly correct text; and (iii) a contamination-robust construction, in which naturally attested misspellings are injected into clean text, so the corrupted passages do not appear verbatim in the source corpora. Because the errors are injected, the gold corrections are exact, and a blind two-annotator study measures how reliably the taxonomy can be applied. Grading is at the level of Thai character clusters, with span-based matching adapted from ERRANT. We evaluate 16 models and find that (1) under the benchmark’s conditions, a model that locates an error almost always fixes it correctly, so what separates models is detection; (2) the evaluated Thai-adapted models do not lead the strongest general models; and (3) in long, mostly correct passages, weaker models accumulate over-edits that pull their scores far below their sentence-level results. On real errors mined from Wikipedia edit history, scores are somewhat lower, but the model ranking largely holds.
%U https://aclanthology.org/2026.aacl-short.17/
%P 211-225
Markdown (Informal)
[ThaiEditBench: A Mechanism-Typed, Length-Stratified Benchmark for Thai Orthographic Error Correction](https://aclanthology.org/2026.aacl-short.17/) (Yongsatianchot et al., AACL-IJCNLP 2026)
ACL