@inproceedings{tran-etal-2026-step,
title = "Step-by-Step {CCG} Parsing with Large Language Models",
author = "Tran, Tu-Anh and
Han, Namgi and
Ishii, Taiga and
Miyao, Yusuke",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1412/",
doi = "10.18653/2026.findings-emnlp.1412",
pages = "30093--30115",
ISBN = "979-8-89176-462-0",
abstract = "We study how step-by-step reasoning should be structured for Combinatory Categorial Grammar (CCG) parsing with Large Language Models (LLMs). Whereas most CCG parsers fix the order of parsing decisions{---}supertagging every word first, then combining the categories bottom-up into a derivation{---}for LLMs that can generate the full derivation as a single output sequence, this order is an open design question. We fine-tune three open-weight models on four reasoning paths along two axes: parsing direction (top-down or bottom-up) and step format (lexically grounded partial derivations or symbolic derivation steps). Our results show that not all reasoning paths improve CCG parsing. On English CCGbank, only the top-down lexically grounded path consistently outperforms a direct, no-reasoning baseline across all three models. In a zero-shot cross-lingual transfer evaluation from English to German, Italian, and Dutch, the same path outperforms a strong CCG parser baseline, despite trailing it on English. Our findings suggest that LLMs may benefit from a different parsing strategy than classical CCG parsers."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tran-etal-2026-step">
<titleInfo>
<title>Step-by-Step CCG Parsing with Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tu-Anh</namePart>
<namePart type="family">Tran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Namgi</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Taiga</namePart>
<namePart type="family">Ishii</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yusuke</namePart>
<namePart type="family">Miyao</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>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>We study how step-by-step reasoning should be structured for Combinatory Categorial Grammar (CCG) parsing with Large Language Models (LLMs). Whereas most CCG parsers fix the order of parsing decisions—supertagging every word first, then combining the categories bottom-up into a derivation—for LLMs that can generate the full derivation as a single output sequence, this order is an open design question. We fine-tune three open-weight models on four reasoning paths along two axes: parsing direction (top-down or bottom-up) and step format (lexically grounded partial derivations or symbolic derivation steps). Our results show that not all reasoning paths improve CCG parsing. On English CCGbank, only the top-down lexically grounded path consistently outperforms a direct, no-reasoning baseline across all three models. In a zero-shot cross-lingual transfer evaluation from English to German, Italian, and Dutch, the same path outperforms a strong CCG parser baseline, despite trailing it on English. Our findings suggest that LLMs may benefit from a different parsing strategy than classical CCG parsers.</abstract>
<identifier type="citekey">tran-etal-2026-step</identifier>
<identifier type="doi">10.18653/2026.findings-emnlp.1412</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1412/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>30093</start>
<end>30115</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Step-by-Step CCG Parsing with Large Language Models
%A Tran, Tu-Anh
%A Han, Namgi
%A Ishii, Taiga
%A Miyao, Yusuke
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F tran-etal-2026-step
%X We study how step-by-step reasoning should be structured for Combinatory Categorial Grammar (CCG) parsing with Large Language Models (LLMs). Whereas most CCG parsers fix the order of parsing decisions—supertagging every word first, then combining the categories bottom-up into a derivation—for LLMs that can generate the full derivation as a single output sequence, this order is an open design question. We fine-tune three open-weight models on four reasoning paths along two axes: parsing direction (top-down or bottom-up) and step format (lexically grounded partial derivations or symbolic derivation steps). Our results show that not all reasoning paths improve CCG parsing. On English CCGbank, only the top-down lexically grounded path consistently outperforms a direct, no-reasoning baseline across all three models. In a zero-shot cross-lingual transfer evaluation from English to German, Italian, and Dutch, the same path outperforms a strong CCG parser baseline, despite trailing it on English. Our findings suggest that LLMs may benefit from a different parsing strategy than classical CCG parsers.
%R 10.18653/2026.findings-emnlp.1412
%U https://aclanthology.org/2026.findings-emnlp.1412/
%U https://doi.org/10.18653/2026.findings-emnlp.1412
%P 30093-30115
Markdown (Informal)
[Step-by-Step CCG Parsing with Large Language Models](https://aclanthology.org/2026.findings-emnlp.1412/) (Tran et al., Findings 2026)
ACL
- Tu-Anh Tran, Namgi Han, Taiga Ishii, and Yusuke Miyao. 2026. Step-by-Step CCG Parsing with Large Language Models. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 30093–30115, Budapest, Hungary. Association for Computational Linguistics.