We present our approach to the AbjadNLP 2026 Arabic Authorship Identification shared task, achieving 4th place. Our key finding is that AraBERT-base (110M) outperforms AraBERT-large (340M) on the test set with macro F1 of 0.8449 versus 0.8096, despite lower validation scores. We handle long passages via sliding window chunking with mean pooling, and use a two-stage classification head with dual dropout for regularization. Per-class analysis reveals that translated works achieve perfect F1 while classical poets remain challenging due to shared formal structures. Our results challenge the "scale is all you need" assumption for stylometric tasks.

This paper describes the system developed by team HCMUS_The Fangs for the AbjadStyleTransfer shared task (ArabicNLP 2026), where we achieved 1st place. We present a contrastive style learning approach for zero-shot Arabic authorship style transfer. Our key discovery is that the 21 test authors-including Nobel laureate Naguib Mahfouz and literary pioneer Taha Hussein-have zero overlap with the 32,784 training authors, transforming this into a pure zero-shot challenge. This insight led us to develop a dual-encoder architecture that learns transferable style representations through contrastive objectives, rather than memorizing author-specific patterns. Our system achieves 19.77 BLEU and 55.74 chrF, outperforming retrieval-augmented generation (+18%) and multi-task learning (+31%). Counter-intuitively, we find that sophisticated architectural modifications like style injection consistently degrade performance, while simpler approaches that preserve pre-trained knowledge excel. Our analysis reveals that for famous authors, pre-trained Arabic language models already encode substantial stylistic knowledge-the key is surfacing it, not learning from scratch.

The rapid advancement of large language mod-els poses significant challenges for content au-thenticity, particularly in under-resourced lan-guages where detection tools remain scarce.We present our winning system for the Abjad-GenEval shared task on Arabic AI-generatedtext detection. Our key insight is that AI-generated text exhibits distinctive patternsacross multiple linguistic levels-from local syn-tax to global semantics-that can be captured bylearning to fuse representations from differenttransformer layers. We introduce aWeightedLayer Poolingmechanism that learns optimallayer combinations, combined withAttentionPoolingfor sequence-level context aggregation.Through systematic experimentation with 15+ approaches, we make a surprising discovery:model architecture selection dominates over so-phisticated training techniques, with DeBERTa-v3 providing +27% relative improvement overAraBERT regardless of training strategy. Oursystem achieves 0.93 F1-score, securing 1st placeamong all participants and outperform-ing the runner-up by 3 absolute points

Adverse Drug Event (ADE) normalization to standardized medical terminologies such as MedDRA presents significant challenges due to lexical and semantic gaps between colloquial user-generated content and formal medical vocabularies. This paper presents our submission to the ALTA 2025 Shared Task on ADE normalization, evaluated using Accuracy@k metrics. Our approach employs distinct methodologies for the development and test phase. In the development phase, we propose a three-stage neural architecture: (1) bi-encoder training to establish semantic representations, (2) lexical-aware fine-tuning to capture morphological patterns alongside semantic similarity, and (3) crossencoder re-ranking for fine-grained discrimination, enabling the model to leverage both distributional semantics and lexical cues through explicit interaction modeling. For the test phase, we utilize the trained bi-encoder from stage (1) for efficient candidate retrieval, then adopt an alternative re-ranking pipeline leveraging large language models with tool-augmented retrieval and multi-stage reasoning. Specifically, a capable model performs reasoning-guided candidate selection over the retrieved top-k results, a lightweight model provides iterative feedback based on reasoning traces, and an automated verification module ensures output correctness with self-correction mechanisms. Our system achieves competitive performance on both development and test benchmarks, demonstrating the efficacy of neural retrieval-reranking architectures and the versatility of LLM-augmented neural pipelines for medical entity normalization tasks.

Recent space missions such as Hubble, Chandra, and JWST have produced a rapidly growing body of scientific literature. Maintaining telescope bibliographies is essential for mission assessment and research traceability, yet current curation processes rely heavily on manual annotation and do not scale. To facilitate progress in this direction, the TRACS @ WASP 2025 shared task provides a benchmark for automatic telescope bibliographic classification based on scientific publications. In this work, we conduct a comparative study of modeling strategies for this task. We first explore traditional machine learning methods such as multinomial Naive Bayes with TF–IDF and CountVectorizer representations. We then evaluate transformer-based multi-label classification using BERT-based scientific language models. Finally, we investigate a task-wise classification approach, where we decompose the problem into separate prediction tasks and train a dedicated model for each. In addition, we experiment with a limited-resource LLM-based approach, showing that even without full fine-tuning and using only a partial subset of the training data, LLMs exhibit promising potential for telescope classification. Our best system achieves a macro F1 of 0.72 with BERT-based models on the test evaluation, substantially outperforming the official openai-gpt-oss-20b baseline (0.31 macro F1).