United States

&quot;Dogwhistles&quot; are expressions intended by the speaker to have two messages: a socially unacceptable &quot;in-group&quot; message understood by a subset of listeners and a benign message intended for the out-group. We take the result of a word-replacement survey of the Swedish population intended to reveal how dogwhistles
are understood, and we show that the difficulty of annotating dogwhistles is reflected in the separability of the space of a sentence-transformer
Swedish BERT trained on general data.

NAACL 2022

Distributional properties of political dogwhistle representations in Swedish
BERT

dogwhistle dataset

semantic modeling

political dogwhistles

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s 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.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

ACL 2025

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

"Dogwhistles" are expressions intended by the speaker to have two messages: a socially unacceptable "in-group" message understood by a subset of listeners and a benign message intended for the out-group. We take the result of a word-replacement survey of the Swedish population intended to reveal how dogwhistles
are understood, and we show that the difficulty of annotating dogwhistles is reflected in the separability of the space of a sentence-transformer
Swedish BERT trained on general data.

workshop paper

## Welcome to NAACL 2022!
The Annual Meeting of the North American Chapter of the Association for Computational Linguistics (NAACL) is taking place July 10-15, 2022 as a hybrid event, in Seattle, WA and online. We are happy to welcome all of you to conference! 
 
The main conference program features oral presentations, in-person and virtual posters and demo sessions, a plenary session for our best paper presentations and awards, keynote presentations, a plenary panel on the place of linguistics and symbolic structures in NLP, and an Industrial Track panel on Careers in NLP.
 
Posters (including Findings of NAACL 2022) and demos are grouped by areas for both the in-person and the virtual sessions. For the virtual component, the talks will be on Zoom and the posters and the demos will be in GatherTown. The Student Research Workshop and will have an oral session and a poster session.
 
The program also features 6 Tutorials and 26 Workshops.
 
 We wish you a wonderful conference!
 
[The NAACL 2022 Organizing Committee](https://2022.naacl.org/committees/organization/)

2022 Annual Conference of the North American Chapter of the Association for Computational Linguistics

Because of the compositionality of natural language, syntactic structure which contains the information about the relationship between words is a key factor for semantic understanding. However, the widely adopted Transformer is hard to learn the syntactic structure effectively in dialogue generation tasks. To explicitly model the compositionaity of language in Transformer Block, we restrict the information flow between words by constructing directed dependency graph and propose Dependency Relation Attention (DRA). Experimental results demonstrate that DRA can further improve the performance of state-of-the-art models for dialogue generation.

ErAConD: Error Annotated Conversational Dialog Dataset for Grammatical Error Correction

Toxic speech is regarded as one of the crucial issues plaguing online social media today. Most recent work on toxic speech detection is constrained to the modality of text and written conversations with very limited work on toxicity detection from spoken utterances or using the modality of speech. In this paper, we introduce a new dataset DeToxy, the first publicly available toxicity annotated dataset for the English language. DeToxy is sourced from various openly available speech databases and consists of over 2 million utterances. We believe that our dataset would act as a benchmark for the relatively new and unexplored SLP task of detecting toxicity from spoken utterances and boost further research in this space. Finally, we also provide strong unimodal baselines for our dataset and compare traditional 2-step cascade and End-to-End (E2E) approaches. Our experiments show that in the case of spoken
utterances, text-based approaches are largely dependent on gold human-annotated transcripts for their performance and also suffer from the problem of keyword bias. However, the presence of speech files in DeToxy helps facilitates the development of E2E speech models which alleviate both the above-stated problems by better capturing speech clues.

DeToxy: A Large-Scale Multimodal Dataset for Toxicity Classification in
Spoken Utterances

A common approach for testing fairness issues in text-based classifiers is through the use of counterfactuals: does the classifier output change if a sensitive attribute in the input is changed? Existing counterfactual generation methods typically rely on wordlists or templates, producing simple counterfactuals that don't take into account grammar, context, or subtle sensitive attribute references, and could miss issues that the wordlist creators had not considered. In this paper, we introduce a task for generating counterfactuals that overcomes these shortcomings, and demonstrate how large language models (LLMs) can be leveraged to make progress on this task. We show that this LLM-based method can produce complex counterfactuals that existing methods cannot, comparing the performance of various counterfactual generation methods on the Civil Comments dataset and showing their value in evaluating a toxicity classifier.

Flexible text generation for counterfactual fairness probing

Lexicons play an important role in content moderation, often being the first line of defense. However, little or no literature exists in analyzing the representation of queer-related words in them. In this paper, we consider twelve well-known English lexicons containing inappropriate words and analyze how gender and sexual minorities are represented in these lexicons. Our analyses reveal that several of these lexicons barely make any distinction between pejorative and non-pejorative queer-related words. We express concern that such unfettered usage of non-pejorative queer-related words may impact queer presence in mainstream discourse. Our analyses further reveal that the lexicons have poor overlap in queer-related words. We finally present a quantifiable measure of consistency and show that several of these lexicons are not consistent in how they include (or omit) queer-related words.

Revisiting Queer Minorities in Lexicons

Despite recent advances in machine learning based hate speech detection, classifiers still struggle with generalizing knowledge to out-
of-domain data samples. In this paper, we investigate the generalization capabilities of deepearning models to different target groups of hate speech under clean experimental settings. Furthermore, we assess the efficacy of three different strategies of unsupervised domain adaptation to improve these capabilities. Given the diversity of hate and its rapid dynamics in the online world (e.g. the evolution of new target groups like virologists during the COVID-19 pandemic), robustly detecting hate aimed at newly identified target groups is a highly relevant research question. We show that naively trained models suffer from a target group specific bias, which can be reduced via domain adaptation. We were able to achieve a relative improvement of the F1-score between 5.8% and 10.7% for out-of-domain target groups of hate speech compared to baseline approaches by utilizing domain adaptation.

Improving Generalization of Hate Speech Detection Systems to Novel Target Groups via Domain Adaptation

Hate Speech has been around in Ethiopia before social media, but with very limited reach. With the coming of social media companies that have no or little business interest to lose in low-resourced languages such as those in Ethiopia, diaspora activists that have nothing or little to lose from engaging in online hate speech, and several technical and institutional challanges, hate speech on social media slowly became mainstream in Ethiopia, tearing societies apart and eventually serving as an animating force for a genocidal war on Tigrayans. In this speech, I will briefly assess the normalization of hate speech in Ethiopia, the factors that led to this, and the role hate speech and social media played during the Tigray war, social media hate speech detection and monitoring, and what should be done going forward. 


Social Media and Hate Speech in time of War: The case of Tigray War

Large pre-trained models are usually fine-tuned on downstream task data, and tested on unseen data.
When the train and test data come from different domains, the model is likely to struggle, as it is not adapted to the test domain. We propose a new approach for domain adaptation (DA), using neuron-level interventions: 
We modify the representation of each test example in specific neurons, resulting in a counterfactual example from the source domain, which the model is more familiar with. The modified example is then fed back into the model.
While most other DA methods are applied during training time, ours is applied during inference only, making it more efficient and applicable. 
Our experiments show that our method improves performance on unseen domains.

IDANI: Inference-time Domain Adaptation via Neuron-level Interventions

Deep learning methods have enabled task oriented
semantic parsing of increasingly complex utterances.
However, a single model is still typically trained and
deployed for each task separately, requiring labeled
training data for each, which makes it challenging to
support new tasks, even within a single business
vertical (e.g., food-ordering or travel booking). In this
paper we describe Cross-TOP (Cross-Schema Task-
Oriented Parsing), a zero-shot method for complex
semantic parsing in a given vertical. By leveraging the
fact that user requests from the same vertical share
lexical and semantic similarities, a single crossschema
parser is trained to service an arbitrary
number of tasks, seen or unseen, within a vertical.
We show that Cross-TOP can achieve high accuracy
on a previously unseen task without requiring any
additional training data, thereby providing a scalable
way to bootstrap semantic parsers for new tasks. As
part of this work we release the FoodOrdering
dataset, a task-oriented parsing dataset in the foodordering
vertical, with utterances and annotations
derived from five schemas, each from a different
restaurant menu.

Cross-TOP: Zero-Shot Cross-Schema Task-Oriented Parsing

Pretrained language models represent the state of the art in NLP, but the successful construction of such models often requires large amounts of data and computational resources. Thus, the paucity of data for low-resource languages impedes the development of robust NLP capabilities for these languages. There has been some recent success in pretraining encoder-only models solely on a combination of low-resource African languages, exemplified by AfriBERTa. In this work, we extend the approach of “small data” pretraining to encoder–decoder models. We introduce AfriTeVa, a family of sequence-to-sequence models derived from T5 that are pretrained on 10 African languages from scratch. With a pretraining corpus of only around 1GB, we show that it is possible to achieve competitive downstream effectiveness for machine translation and text classification, compared to larger models trained on much more data. All the code and model checkpoints described in this work are publicly available at https://github.com/castorini/
afriteva

AfriTeVa: Extending “Small Data” Pretraining Approaches to Sequence-to-Sequence Models

Knowledge Graphs (KGs) are directed labeled graphs representing entities and the relationships between them. Most prior work focuses on supervised or semi-supervised approaches which require large amounts of annotated data. While unsupervised approaches do not need labeled training data, most existing methods either generate too many redundant relations or require manual mapping of the extracted relations to a known schema. To address these limitations, we propose an unsupervised method for KG generation that requires neither labeled data nor manual mapping to the predefined relation schema. Instead, our method leverages sentence-level semantic similarity for automatically generating relations between pairs of entities. Our proposed method outperforms two baseline systems when evaluated over four datasets.

Downloads

Next from NAACL 2022

ErAConD: Error Annotated Conversational Dialog Dataset for Grammatical Error Correction

Similar lecture

Who leads? Who follows? Temporal dynamics of political dogwhistles in Swedish online communities

.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 NAACL 2022

ErAConD: Error Annotated Conversational Dialog Dataset for Grammatical Error Correction

Similar lecture

Who leads? Who follows? Temporal dynamics of political dogwhistles in Swedish online communities

Downloads