United States

Human annotation is a foundational component of modern natural language processing (NLP). Labeled datasets underpin widely used benchmarks for sentiment analysis, toxicity detection, hate speech classification, and stance detection. Within standard NLP workflows, annotation is generally treated as a technical process aimed at recovering an objective ground truth according to predefined guidelines. This paper argues that such a view overlooks the inherently interpretive nature of annotation. Drawing on insights from sociolinguistics, discourse analysis, and cultural theory, and on a growing empirical literature on annotator subjectivity, we propose that annotation should be understood as a culturally situated interpretive practice. Annotators rely on culturally shaped norms, values, and communicative expectations when interpreting linguistic meaning, and labels in NLP datasets often reflect culturally specific interpretations rather than universal truths. We position this argument relative to recent work on perspectivism, annotator-aware modeling, and cross-cultural annotation, and we use published findings from large-scale cross-cultural annotation studies to illustrate the concrete consequences of treating annotation as objective. We close with a research agenda for culturally informed annotation practice that includes operational recommendations on documentation, modeling, and evaluation.

ACL 2026

Annotation as Cultural Interpretation: Rethinking Data Labeling in NLP

cultural

sociolinguistics

annotation

It is my great pleasure and honor to welcome you to the 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026), held in San Diego, California, USA, July 2–7, 2026. ACL 2026 continues our fields tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.

It is exciting to see the many developments in the field reflected at our conference, such as the advancement of large language models (born out of our community) into many application directions. This is not without challenges, partly due to unsolved problems, and partly due to misuse of the technology. All this is reflected well at ACL 2026. The conference also reflects that the field is not only about advancing technology, but also the science of fundamentally understanding language and how it is use by many different speaker communities. <br>
[Full Message from General Chair Philipp Koehn](https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing) <br><br><br>
<html><button style="display: inline-flex; align-items: center; justify-content: center; white-space: nowrap; border-radius: 9999px; font-weight: bold; background: #7c3aed; color: white; font-family: 'Space Grotesk', sans-serif; height: 40px; font-size: 16px; padding: 0 20px; border: none; cursor: pointer" onclick="window.open('https://drive.google.com/file/d/1QiJFOmMjvDgsDo3i1XzlQnImxe3RU_Cy/view?usp=sharing')">Conference Handbook</button></html>
<br><br><br><br>

To access this event page, you need to log in with the **email address you registered with**. <br>Access credentials will be sent to your email from Underline -  subject line "Welcome to ACL 2026". Please be sure to check your spam email folder if you do not see an email confirmation right away.

To access this event page, you are required to register.
Please complete your registration to continue.

We recommend reading [**the registration information**](https://2026.aclweb.org/registration/) first.

Registration Required

The 64th Annual Meeting of the Association for Computational Linguistics (ACL 2026) will take place in San Diego, California, from July 2 - 7, 2026.

workshop paper

Large language models (LLMs) are increasingly used for mental health applications, raising questions about whether they reflect established clinical knowledge. Clinical psychology has documented systematic cultural differences in the presentation of depression symptoms, with Western populations emphasizing emotional symptoms and many East Asian populations reporting more somatic symptoms. We evaluate whether general-purpose LLMs reproduce these clinically established cross-cultural patterns using prompts grounded in clinical descriptions of depression. We examine model responses under different cultural personas and languages. We find that LLMs struggle to reproduce expected cultural patterns when prompted in English. Prompting in major Eastern languages improves alignment in some configurations, suggesting that language cues partially activate cultural knowledge. However, model behavior remains dominated by a strong, culture-invariant hierarchy of depression symptoms that often overrides cultural cues, highlighting limitations in current LLMs for mental health applications.

Somatic in the East, Psychological in the West?: A Clinically-Grounded Evaluation of Cross-Cultural Depression Symptoms in LLMs

AI systems for mental health are developed predominantly using data from Western, Educated, Industrialized, Rich, and Democratic (WEIRD) populations, raising concerns about their validity, fairness, and generalizability across geo-cultural contexts. This limitation is especially consequential in mental health, where linguistic expression, symptom presentation, help-seeking behavior, and access to care vary substantially across populations. We argue that culturally responsive AI mental health systems require explicit attention to culture throughout the development lifecycle, from data collection and annotation to evaluation, deployment, and governance. We present a ten-part sociotechnical framework for developing culturally responsive AI mental health applications to provide AI researchers and practitioners with an actionable roadmap for building more equitable, reliable, and contextually appropriate mental health technologies.

Beyond WEIRD Data: A Sociotechnical Framework for Culturally Responsive AI in Mental Health

Wildfire management systems increasingly rely on computational models and centralized policy frameworks for fire prediction and mitigation, yet these approaches overlook localized ecological knowledge developed through long-term interaction with landscapes. Traditional Ecological Knowledge (TEK), developed and practiced by Indigenous communities over generations, contains detailed operational practices for controlled burning and community-based fire governance, but remains under-integrated into computational decision pipelines. We propose a knowledge-grounded framework that converts narrative TEK into structured operational claims, and enables Language Models (LMs) to generate wildfire management recommendations grounded in documented Indigenous practices. Our pipeline extracts verifiable knowledge claims and uses citation-based reasoning to produce recommendations. Through case studies of three Indigenous communities, the Jemez Pueblo of North America, Karen of Myanmar, and Pintupi of Australia, we show that the system provides grounded guidance for wildfire scenarios, while reliably rejecting queries beyond the available knowledge base.

Indigenous Inheritances: Integrating Traditional Ecological Knowledge into Wildfire Mitigation

Code-switching is often modeled in NLP as a structural or token-level phenomenon, overlooking its role as a discourse practice shaped by social and cultural context. In this work, we propose topic-based annotation as a framework for analyzing cultural and subcultural variation in bilingual discourse. Using large language models, we annotate 3,691 code-switched sentences from Spanish-English (Miami) and Spanish-Guaraní (Paraguay) corpora with topic and discourse-level information, integrating sociolinguistic metadata. Our analysis reveals systematic relationships between discourse topics, language choice, and social variables such as gender and language dominance. We observe subcultural variation within the Miami community and a clear diglossic distribution in Paraguay, where Guaraní is associated with formal domains and Spanish with informal communication. These findings suggest that modeling code-switching through discourse-level categories provides a more complete representation of multilingual communication and enables both cross-cultural and intra-cultural comparison at scale.

Modeling Cultural and Subcultural Variation in Code-Switched Discourse with Topic Annotation

Adapting large language models (LLMs) to extremely low-resource languages remains challenging due to severe data scarcity and the lack of structured linguistic supervision. We introduce GCCLA, a graph-conditioned cross-lingual adaptation framework that integrates multilingual knowledge graphs into parameter-efficient LLM adaptation. GCCLA conditions a frozen multilingual LLM on structured semantic and typological relations encoded in a multilingual graph, providing a strong inductive bias for data-efficient transfer. We instantiate and evaluate the framework through a focused case study on English-to-Amharic-to-Tigrinya transfer, where labeled data is extremely limited. By separating knowledge representation from language modeling, GCCLA stabilizes learning and improves sample efficiency in few-shot regimes. We evaluate the approach on five tasks, sentiment analysis, named entity recognition, natural language inference, question answering, and extractive summarization, under extreme data scarcity, with as few as 0–1000 labeled Tigrinya examples. Experimental results show that GCCLA consistently outperforms multilingual, translation-based, and parameter-efficient baselines, achieves competitive performance with as few as 100 labeled examples, and degrades gracefully under partial graph coverage. These findings demonstrate that graph conditioning is an effective principle for data-efficient cross-lingual adaptation of LLMs advancing equitable NLP.

GCCLA: Graph-Conditioned Cross-Lingual Adaptation of Large Language Models Under Extreme Data Scarcity (A Case Study in Tigrigna)

The global use of artificial intelligence for content creation has led to an interest in assessing the ability to generate culturally representative content. We explore whether generated content is localized through cultural markers that are (a) inserted within a generic narrative template or (b) distributed throughout to create culturally-salient narratives, plots, and themes. We analyze stories generated by five models on 125 topics for 193 nationalities. We find that only a small subset (9-17%) of the vocabulary accounts for the variation in stories across nationalities. After removal of these cultural markers, we find that the remaining narrative is homogeneous across cultural contexts. Moreover, we find varying degrees of stereotypicality in cultural markers. Finally, markers from 19 countries, mostly located in the Global South where dominant languages are lower resourced, are on average offensive, while those from the rest are on average non-offensive or neutral.

Cultural Markers and Narrative Homogeneity in AI-Generated Stories

Although artificial intelligence (AI) and Human-Computer Interaction (HCI) systems are often presented as universal solutions, their design remains predominantly text-first, underserving primarily oral languages and indigenous communities. This position paper uses Guaraní, an official and widely spoken language of Paraguay, as a case study to argue that language support in AI remains insufficient unless it aligns with lived oral practices. We propose an alternative to the standard "text-to-speech'' pipeline, proposing instead an oral-first multi-agent architecture. The architecture is intended to illustrate the design implications of our cultural argument rather than to propose a deployable technical solution. By decoupling Guaraní natural language understanding from dedicated agents for conversation state and community-led governance, we sketch a technical framework that respects indigenous data sovereignty and diglossia. Our work moves beyond mere recognition to focus on turn-taking, repair, and shared context as the primary locus of interaction. We conclude that for AI to be truly culturally grounded, it must shift from adapting oral languages to text-centric systems to treating spoken conversation as a first-class design requirement, ensuring digital ecosystems empower rather than overlook diverse linguistic practices.

Let’s Talk, Not Type: An Oral-First Multi-Agent Architecture for Guaraní

We evaluate whether open-source LLMs can produce proficiency-graded English adaptations of entries from the \textit{Diccionario de colombianismos} (DiCol), a Colombian Spanish lexicographic resource used in language teaching. Three 7--8B instruction-tuned models---Llama~3.1, Qwen2.5, and Mistral---generate Beginner, Intermediate, and Advanced translations for all 8{,}252 definitions using structured zero-shot prompts identical across levels except for the target CEFR band. Automated metrics show that Intermediate targeting collapses (73--83\% classified as Advanced by vocabulary, $\chi^2 > 705$, $p < .001$) and that Advanced outputs expand 4.9--8.2$\times$ relative to the source. Expert annotation of a 360-entry stratified sample ($\kappa = 0.61$--$0.68$) identifies hallucination in 19\% of entries (Fleiss' $\kappa = 0.77$ for cultural preservation categories, 97\% unanimity among flagged cases). Hallucination concentrates in the Advanced condition (81\%, $\chi^2 = 86.6$, $p < .001$) and is associated with higher expansion ($U = 16{,}662$, $p < .001$, $r = 0.68$), manifesting primarily as generic elaboration and, in a smaller proportion, as Colombia-stereotyping and pragmatic polarity inversion. We discuss these findings through the lens of \citeauthor{venuti1995invisibility}'s domestication framework and describe the observed pattern as \emph{algorithmic domestication}.

LLM-Adapted Colombian Spanish Lexicography: Proficiency Control, Hallucination, and Cultural Distortion

Large Language Models (LLMs) show unbalanced knowledge of cultures across the globe, favoring high-resource cultures over low-resource ones. A possible way to tackle this issue is to fine-tune LLMs on culturally specific data. However, fine-tuning recent LLMs requires high computational resources as well as memory storage, which triggered the development of parameter-efficient fine-tuning (PEFT) approaches, the most widespread being LoRA. In this article, we investigate the use of another class of PEFT approaches, namely soft prompt methods (prompt-tuning and prefix-tuning), to improve LLMs' cultural knowledge across diverse cultures. We focus on cultural alignment on Multiple-Choice Questions of cultural commonsense knowledge. On this task with limited fine-tuning data, we show that soft-prompt-based methods outperform LoRA in comparable settings. Moreover, the trained soft prompts are interpretable and capture similarities between cultures.

Soft Prompts for Adapting LLMs to Cultural Commonsense Knowledge

Culture shapes the objects people use and for what purposes, yet mainstream Vision-Language (VL) datasets frequently exhibit cultural biases, disproportionately favoring higher-income, Western contexts. This imbalance reduces model generalizability and perpetuates performance disparities, especially impacting lower-income and non-Western communities. To address these disparities, we propose a novel function-centric framework that categorizes objects by the functions they fulfill, across diverse cultural and economic contexts. We implement this framework by creating the Culture Affordance Atlas, a re-annotated and culturally grounded restructuring of the Dollar Street dataset spanning 46 functions and 288 objects. Through extensive empirical analyses using the CLIP model, we demonstrate that function-centric labels substantially reduce socioeconomic performance gaps between high- and low-income groups by a median of 6 pp (statistically significant), improving model effectiveness for lower-income contexts. Furthermore, our analyses reveals numerous culturally essential objects that are frequently overlooked in prominent VL datasets. Our contributions offer a scalable pathway toward building inclusive VL datasets and equitable AI systems.

Downloads

Next from ACL 2026

Somatic in the East, Psychological in the West?: A Clinically-Grounded Evaluation of Cross-Cultural Depression Symptoms in LLMs

Similar lecture

Lost in Dialect: The Annotation Gap in Multilingual LLM Safety

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

.css-70qvj9{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}Downloads

Next from ACL 2026

Somatic in the East, Psychological in the West?: A Clinically-Grounded Evaluation of Cross-Cultural Depression Symptoms in LLMs

Similar lecture

Lost in Dialect: The Annotation Gap in Multilingual LLM Safety

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES

Downloads