Human annotation is the empirical foundation of much NLP research, from dataset construction to model evaluation, but papers often leave unclear who produced the annotations and how the annotation process was controlled. We provide the first large-scale, task-level audit of human annotation reporting across major NLP venues, asking which annotation details are documented, which are missing, and how reporting varies across time, topic, venue, and intended use of human judgment. We introduce a unified taxonomy of annotation-reporting practices and validate an LLM-assisted extraction pipeline against Annotated_gold, a human-adjudicated gold standard of 41 papers and 72 annotation tasks, where the best model reaches human-comparable agreement with adjudicated labels, with Krippendorff’s alpha = 0.606 versus 0.585 for human–human agreement. Using this pipeline, we construct Annotated_llm, a dataset covering ACL-venue papers from 2018–2025, with 2,667 extracted annotation tasks from 1,603 papers, and find that papers frequently report operational details such as recruitment strategies, annotator expertise, and annotation volume, but often omit details needed to assess annotation validity, including training, language proficiency, compensation, socio-demographics, adjudication, and agreement values, especially in model-evaluation studies. Our results show that annotation reporting in NLP has improved over time but remains uneven, and they establish a scalable framework and bare-minimum reporting recommendations for making human annotation more reliable, reproducible, and interpretable.

Emotions play a fundamental role in the decision-making process, shaping human actions across diverse disciplines. The extensive usage of emotion intensity detection approaches has generated substantial research interest during the last few years. Efficient multi-label emotion intensity detection remains unsatisfactory even for high-resource languages, with a substantial performance gap among well-resourced and under-resourced languages. Team {textbf{Tewodros}} participated in SemEval-2025 Task 11, Track B, focusing on detecting text-based emotion intensity. Our work involved multi-label emotion intensity detection across three languages: Amharic, English, and Spanish, using the (afro-xlmr-large-76L), (DeBERTa-v3-base), and (BERT-base-Spanish-wwm-uncased) models. The models achieved an average F1 score of 0.6503 for Amharic, 0.5943 for English, and an accuracy score of 0.6228 for Spanish. These results demonstrate the effectiveness of our models in capturing emotion intensity across multiple languages.

In this paper, we describe the system developed by Team EM-26 for the Shared Task on Caste and Migration Hate Speech Detection at LTEDI@LDK 2025. The task addresses the challenge of recognizing caste-based and migration related hate speech in Tamil social media text, a language that is both nuanced and under resourced for machine learning. To tackle this, we fine-tuned the multilingual transformer XLM-RoBERTa-Large on the provided training data, leveraging its cross-lingual strengths to detect both explicit and implicit hate speech. To improve performance, we applied social media focused preprocessing techniques, including Tamil text normalization and noise removal. Our model achieved a macro F1-score of 0.6567 on the test set, highlighting the effectiveness of multilingual transformers for low resource hate speech detection. Additionally, we discuss key challenges and errors in Tamil hate speech classification, which may guide future work toward building more ethical and inclusive AI systems. The source code is available on GitHub.1

Social media platforms and user-generated content, such as tweets, comments, and blog posts often contain offensive language, including racial hate speech, personal attacks, and sexual harassment. Detecting such inappropriate language is essential to ensure user safety and to prevent the spread of hateful behavior and online aggression. Approaches base on conventional machine learning and deep learning have shown robust results for high-resource languages like English and find it hard to deal with code-mixed text, which is common in bilingual communication. We participated in the shared task “LT-EDI@LDK 2025” organized by DravidianLangTech, applying the BERT-base multilingual cased model and achieving an F1 score of 0.63. These results demonstrate how our model effectively processes and interprets the unique linguistic features of code-mixed content. The source code is available on GitHub.1

As machine-generated texts (MGT) become increasingly similar to human writing, these dis- tinctions are harder to identify. In this paper, we as the CIC-NLP team present our submission to the Gen-AI Content Detection Workshop at COLING 2025 for Task 1 Subtask A, which involves distinguishing between text generated by LLMs and text authored by humans, with an emphasis on detecting English-only MGT. We applied the DistilBERT model to this binary classification task using the dataset provided by the organizers. Fine-tuning the model effectively differentiated between the classes, resulting in a micro-average F1-score of 0.70 on the evaluation test set. We provide a detailed explanation of the fine-tuning parameters and steps involved in our analysis.

Machine-written texts are gradually becoming indistinguishable from human-generated texts, leading to the need to use sophisticated methods to detect them. Team CIC-NLP presents work in the Gen-AI Content Detection Task 1 at COLING 2025 Workshop: the focus of our work is on Subtask B of Task 1, which is the classification of text written by machines and human authors, with particular attention paid to identifying multilingual binary classification problem. Usng mBERT, we addressed the binary classification task using the dataset provided by the GenAI Detection Task team. mBERT acchieved a macro-average F1-score of 0.72 as well as an accuracy score of 0.73.

AI-generated text now matches human writing so well that telling them apart is very difficult. Our CIC-NLP team submits results for the DravidianLangTech@NAACL 2025 shared task to reveal AI-generated product reviews in Dravidian languages. We performed a binary classification task with XLM-RoBERTa-Base using the DravidianLangTech@NAACL 2025 datasets offered by the event organizers. Through training the model correctly, our tests could tell between human and AI-generated reviews with scores of 0.96 for Tamil and 0.88 for Malayalam in the evaluation test set. This paper presents detailed information about preprocessing, model architecture, hyperparameter fine-tuning settings, the experimental process, and the results. The source code is available on GitHub1.

Misinformation is a growing problem for technologycompanies and for society. Although there exists a large body of related work on identifying fake news in predominantlyresource languages, there is unfortunately a lack of such studies in low-resource languages (LRLs). Because corpora and annotated data are scarce in LRLs, the identification of false information remains at an exploratory stage. Fake news detection is critical in this digital era to avoid spreading misleading information. This research work presents an approach to Detect Fake News in Dravidian Languages. Our team CIC-NLP work primarily targets Task 1 which involves identifying whether a given social platform news is original or fake. For fake news detection (FND) problem, we used mBERT model and utilized the dataset that was provided by the organizers of the workshop. In this work, we describe our findings and the results of the proposed method. Our mBERT model achieved an F1 score of 0.853.

This study goes into our team’s active participation in the Hate and Offensive Language Detection in Telugu Codemixed Text (HOLDTelugu) shared task, which is an essential component of the DravidianLangTech@EACL 2024 workshop. The ultimate goal of this collaborative work is to push the bounds of hate speech recognition, especially tackling the issues given by codemixed text in Telugu, where English blends smoothly. Our inquiry offers a complete evaluation of the task’s aims, the technique used, and the precise achievements obtained by our team, providing a full insight into our contributions to this crucial linguistic and technical undertaking.