@inproceedings{ranjit-etal-2026-coda,
title = "{C}o{DA}: A Human-{LM} Collaborative Codebook Development Assistant for Domain-Specific Text Annotation",
author = "Ranjit, Jaspreet and
Chellappa, Swaminathan and
Costante, Augusto Rivas and
Manjunath, Chandan and
Onasch-Vera, Laura and
Nadel, Hailey Winetrobe and
Rice, Eric and
Swayamdipta, Swabha",
editor = "Agrawal, Sweta and
Abad, Alberto",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-demos.53/",
pages = "547--557",
ISBN = "979-8-89176-461-3",
abstract = "We present CoDA, an open-source human-LM collaborative codebook development assistant that helps domain experts develop annotation guidelines for annotating new variables in unstructured text with language models. Existing language model annotation systems primarily focus on downstream classification performance, while assuming the availability of high-quality annotation guidelines, or codebooks. In practice, developing these codebooks is time-consuming and challenging, particularly for concepts that require expert knowledge. CoDA addresses this gap by collaborating with domain experts to iteratively develop codebooks through targeted review of incorrect language model predictions and incorporating expert feedback into the evolving codebook. We demonstrate CoDA on two real-world case studies: extracting health indicators from homelessness case management notes and classifying sentiment of wildlife trafficking posts on social media. Across both domains, CoDA produces codebooks that achieve downstream classification performance comparable to expert codebooks while reducing codebook development time from several weeks to two hours and requiring 5$\times$ fewer manually annotated examples. Applying CoDA to generate annotations at scale, we identified 2,410 case notes that contained health discussions or referrals and 10,196 social media posts expressing negative sentiment that may warrant investigation for potential wildlife trafficking. By accelerating codebook development, CoDA helps practitioners use language models to extract domain-specific insights from unstructured text at scale. making in high-impact domains."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ranjit-etal-2026-coda">
<titleInfo>
<title>CoDA: A Human-LM Collaborative Codebook Development Assistant for Domain-Specific Text Annotation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jaspreet</namePart>
<namePart type="family">Ranjit</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Swaminathan</namePart>
<namePart type="family">Chellappa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Augusto</namePart>
<namePart type="given">Rivas</namePart>
<namePart type="family">Costante</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chandan</namePart>
<namePart type="family">Manjunath</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Laura</namePart>
<namePart type="family">Onasch-Vera</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hailey</namePart>
<namePart type="given">Winetrobe</namePart>
<namePart type="family">Nadel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eric</namePart>
<namePart type="family">Rice</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Swabha</namePart>
<namePart type="family">Swayamdipta</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 2026 Conference on Empirical Methods in Natural Language Processing: System Demonstrations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sweta</namePart>
<namePart type="family">Agrawal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alberto</namePart>
<namePart type="family">Abad</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">979-8-89176-461-3</identifier>
</relatedItem>
<abstract>We present CoDA, an open-source human-LM collaborative codebook development assistant that helps domain experts develop annotation guidelines for annotating new variables in unstructured text with language models. Existing language model annotation systems primarily focus on downstream classification performance, while assuming the availability of high-quality annotation guidelines, or codebooks. In practice, developing these codebooks is time-consuming and challenging, particularly for concepts that require expert knowledge. CoDA addresses this gap by collaborating with domain experts to iteratively develop codebooks through targeted review of incorrect language model predictions and incorporating expert feedback into the evolving codebook. We demonstrate CoDA on two real-world case studies: extracting health indicators from homelessness case management notes and classifying sentiment of wildlife trafficking posts on social media. Across both domains, CoDA produces codebooks that achieve downstream classification performance comparable to expert codebooks while reducing codebook development time from several weeks to two hours and requiring 5\times fewer manually annotated examples. Applying CoDA to generate annotations at scale, we identified 2,410 case notes that contained health discussions or referrals and 10,196 social media posts expressing negative sentiment that may warrant investigation for potential wildlife trafficking. By accelerating codebook development, CoDA helps practitioners use language models to extract domain-specific insights from unstructured text at scale. making in high-impact domains.</abstract>
<identifier type="citekey">ranjit-etal-2026-coda</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-demos.53/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>547</start>
<end>557</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CoDA: A Human-LM Collaborative Codebook Development Assistant for Domain-Specific Text Annotation
%A Ranjit, Jaspreet
%A Chellappa, Swaminathan
%A Costante, Augusto Rivas
%A Manjunath, Chandan
%A Onasch-Vera, Laura
%A Nadel, Hailey Winetrobe
%A Rice, Eric
%A Swayamdipta, Swabha
%Y Agrawal, Sweta
%Y Abad, Alberto
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: System Demonstrations
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-461-3
%F ranjit-etal-2026-coda
%X We present CoDA, an open-source human-LM collaborative codebook development assistant that helps domain experts develop annotation guidelines for annotating new variables in unstructured text with language models. Existing language model annotation systems primarily focus on downstream classification performance, while assuming the availability of high-quality annotation guidelines, or codebooks. In practice, developing these codebooks is time-consuming and challenging, particularly for concepts that require expert knowledge. CoDA addresses this gap by collaborating with domain experts to iteratively develop codebooks through targeted review of incorrect language model predictions and incorporating expert feedback into the evolving codebook. We demonstrate CoDA on two real-world case studies: extracting health indicators from homelessness case management notes and classifying sentiment of wildlife trafficking posts on social media. Across both domains, CoDA produces codebooks that achieve downstream classification performance comparable to expert codebooks while reducing codebook development time from several weeks to two hours and requiring 5\times fewer manually annotated examples. Applying CoDA to generate annotations at scale, we identified 2,410 case notes that contained health discussions or referrals and 10,196 social media posts expressing negative sentiment that may warrant investigation for potential wildlife trafficking. By accelerating codebook development, CoDA helps practitioners use language models to extract domain-specific insights from unstructured text at scale. making in high-impact domains.
%U https://aclanthology.org/2026.emnlp-demos.53/
%P 547-557
Markdown (Informal)
[CoDA: A Human-LM Collaborative Codebook Development Assistant for Domain-Specific Text Annotation](https://aclanthology.org/2026.emnlp-demos.53/) (Ranjit et al., EMNLP 2026)
ACL
- Jaspreet Ranjit, Swaminathan Chellappa, Augusto Rivas Costante, Chandan Manjunath, Laura Onasch-Vera, Hailey Winetrobe Nadel, Eric Rice, and Swabha Swayamdipta. 2026. CoDA: A Human-LM Collaborative Codebook Development Assistant for Domain-Specific Text Annotation. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 547–557, Budapest, Hungary. Association for Computational Linguistics.