The identification of narrative similarity is a complex NLP challenge that requires modeling deeper plot and thematic alignment rather than relying solely on lexical overlap. In this paper, we detail the participation of team NLP-FSDM in SemEval-2026 Task 4. Our approach utilizes the bge-large-en-v1.5 encoder. For Track A, we fine-tune it using Multiple Negatives Ranking Loss (MNRL), while for Track B we rely on the pretrained encoder to generate fixed narrative representations. We achieved an accuracy of 65.50% in Track A and 62.50% in Track B. This paper provides an extensive comparison of our results with competitive baselines and top-performing systems, analyzing the efficacy of dense encoders in low-resource narrative contexts.

This paper describes the NLP-FSDM system for SemEval-2026 Task 2, Subtask 1 on longitudinal affect assessment. The task requires predicting Valence and Arousal (V & A) scores for sequences of ecological essays and feeling words written over time. We adopt ModernBERT-large as a text encoder and formulate the task as a joint regression problem optimized using a Concordance Correlation Coefficient (CCC) loss combined with a lightly weighted Mean Absolute Error (MAE) term. To reduce variance induced by fine-tuning large transformers on relatively small user-specific datasets, we employ a three-seed ensemble. Finally, we introduce a lightweight post-inference temporal smoothing mechanism applied per user to improve within-user consistency. Our system achieves an rcomposite of 0.546 for Valence and 0.453 for Arousal, demonstrating stable cross-dimensional performance without explicitly modeling sequential dependencies.

The generation of highly fluent text by Large Language Models (LLMs) poses a significant challenge to information integrity and academic research. In this paper, we introduce the Multi-Domain Detection of AI-Generated Text (M-DAIGT) shared task, which focuses on detecting AI-generated text across multiple domains, particularly in news articles and academic writing. M-DAIGT comprises two binary classification subtasks: News Article Detection (NAD) (Subtask 1) and Academic Writing Detection (AWD) (Subtask 2). To support this task, we developed and released a new large-scale benchmark dataset of 30,000 samples, balanced between human-written and AI-generated texts. The AI-generated content was produced using a variety of modern LLMs (e.g., GPT-4, Claude) and diverse prompting strategies. A total of 46 unique teams registered for the shared task, of which four teams submitted final results. All four teams participated in both Subtask 1 and Subtask 2. We describe the methods employed by these participating teams and briefly discuss future directions for M-DAIGT.

We present an overview of the AraGenEval shared task, organized as part of the ArabicNLP 2025 conference. This task introduced the first benchmark suite for Arabic authorship analysis, featuring three subtasks: Authorship Style Transfer, Authorship Identification, and AI-Generated Text Detection. We curated high-quality datasets, including over 47,000 paragraphs from 21 authors and a balanced corpus of human- and AI-generated texts. The task attracted significant global participation, with 72 registered teams from 16 countries. The results highlight the effectiveness of transformer-based models, with top systems leveraging prompt engineering for style transfer, model ensembling for authorship identification, and a mix of multilingual and Arabic-specific models for AI text detection. This paper details the task design, datasets, participant systems, and key findings, establishing a foundation for future research in Arabic stylistics and trustworthy NLP.

This paper presents the application of BERT inSemEval 2024 Task 2, Safe Biomedical Natu-ral Language Inference for Clinical Trials. Themain objectives of this task were: First, to in-vestigate the consistency of BERT in its rep-resentation of semantic phenomena necessaryfor complex inference in clinical NLI settings.Second, to investigate the ability of BERT toperform faithful reasoning, i.e., make correctpredictions for the correct reasons. The submit-ted model is fine-tuned on the NLI4CT dataset,which is enhanced with a novel contrast set,using binary cross entropy loss.

This paper presents our system and findings for SemEval 2024 Task 1 Track A Supervised Semantic Textual Relatedness. The main objective of this task was to detect the degree of semantic relatedness between pairs of sentences. Our submitted models (ranked 6/24 in Algerian Arabic, 7/25 in Spanish, 12/23 in Moroccan Arabic, and 13/36 in English) consist of various transformer-based models including MARBERT-V2, mDeBERTa-V3-Base, DarijaBERT, and DeBERTa-V3-Large, fine-tuned using different loss functions including Huber Loss, Mean Absolute Error, and Mean Squared Error.

This paper presents our systems and findings for SemEval-2023 Task 12: AfriSenti-SemEval: Sentiment Analysis for Low-resource African Languages. The main objective of this task was to determine the polarity of a tweet (positive, negative, or neutral). Our submitted models (highest rank is 1 and lowest rank is 21 depending on the target Track) consist of various Transformer-based approaches.

This paper presents our system and findings for SemEval 2023 Task 9 Tweet Intimacy Analysis. The main objective of this task was to predict the intimacy of tweets in 10 languages. Our submitted model (ranked 28/45) consists of a transformer-based approach with data augmentation via machine translation.

This paper presents our proposed method for english documents genre classification in the context of SemEval 2023 task 3, subtask 1. Our method use ensemble technique to combine four distinct models predictions: Longformer, RoBERTa, GCN, and a sentences number-based model. Each model is optimized on simple objectives and easy to grasp. We provide snippets of code that define each model to make the reading experience better. Our method ranked 12th in documents genre classification for english texts.

In this paper, we present our system and findings for SemEval-2022 Task 6 - iSarcasmEval: Intended Sarcasm Detection in English. The main objective of this task was to identify sarcastic tweets. This task was challenging mainly due to (1) the small training dataset that contains only 3468 tweets and (2) the imbalanced class distribution (25% sarcastic and 75% non-sarcastic). Our submitted model (ranked eighth on Sub-Task A and fifth on Sub-Task C) consists of a Transformer-based approach (BERTweet model).

This paper presents our proposed methods for the iSarcasmEval shared task. The shared task consists of three different subtasks. We participate in both subtask A and subtask C. The purpose of subtask A was to predict if a text is sarcastic while the aim of subtask C is to determine which text is sarcastic given a sarcastic text and its non-sarcastic rephrase. Both of the developed solutions used BERT pre-trained models. The proposed models are optimized on simple objectives and are easy to grasp. However, despite their simplicity, our methods ranked 4 and 2 in iSarcasmEval subtask A and subtask C for Arabic texts.

Toxic spans detection is an emerging challenge that aims to find toxic spans within a toxic text. In this paper, we describe our solutions to tackle toxic spans detection. The first solution, which follows a supervised approach, is based on SpanBERT model. This latter is intended to better embed and predict spans of text. The second solution, which adopts an unsupervised approach, combines linear support vector machine with the Local Interpretable Model-Agnostic Explanations (LIME). This last is used to interpret predictions of learning-based models. Our supervised model outperformed the unsupervised model and achieved the f-score of 67,84% (ranked 22/85) in Task 5 at SemEval-2021: Toxic Spans Detection.

AraBERT is an Arabic version of the state-of-the-art Bidirectional Encoder Representations from Transformers (BERT) model. The latter has achieved good performance in a variety of Natural Language Processing (NLP) tasks. In this paper, we propose an effective AraBERT embeddings-based method for dealing with offensive Arabic language in Twitter. First, we pre-process tweets by handling emojis and including their Arabic meanings. To overcome the pretrain-finetune discrepancy, we substitute each detected emojis by the special token [MASK] into both fine tuning and inference phases. Then, we represent tweets tokens by applying AraBERT model. Finally, we feed the tweet representation into a sigmoid function to decide whether a tweet is offensive or not. The proposed method achieved the best results on OffensEval 2020: Arabic task and reached a macro F1 score equal to 90.17%.