Emotional tone plays a central role in persuasion, yet its impact on computational assessments of political argument quality in real world election campaign speeches remains understudied. In this work, we investigate whether positive emotional framing correlates with higher perceived convincingness in political arguments. We fine-tune language models on argument quality datasets and test their ability to transfer convincingness predictions to real-world campaign speeches. Using a corpus of U.S. presidential campaign speeches, we analyze emotional polarity in relation to predicted persuasive strength to test whether positively framed arguments are judged more convincing than neutral or negative ones. Our empirical analysis shows that political parties rely heavily on argumentation during their election campaigns. Also, we found the evidence that politicians strategically employ emotional cues within their arguments during these campaign speeches, with positive emotions being more strongly associated with persuasive strength, for example in topics such as USMCA’s Effect on American Jobs and Agriculture, Border Control Policies, Progressive Tax Reforms. At the same time, we find that negative emotions have a weaker yet still non-negligible influence on voter persuasion in topics such as City Crime and Civil Unrest and White Supremacist Violence (Charlottesville Incident).

Arabic sentiment analysis is challenged by morphological complexity and lexical variation across Arabic dialects, compounded by subjectivity in how speakers and writers express sentiment. In this paper, we present our submission for the AraSentEval 2026 Shared Task on Arabic Dialect Sentiment Analysis. We propose SARF (صرف) a multi-view architectural framework that integrates surface-level context with stemmed and rooted morphological perspectives using a shared MARBERTv2 encoder. Our system employs a hybrid BERT-CNN-BiLSTM-Attention architecture to capture both local sentiment n-grams and global sequential dependencies. Experimental results show that while individual morphological normalization strategies (stemming or rooting) may degrade performance, their joint integration via cross-morphological attention provides robust features across diverse dialects. Our final system achieved a competitive macro-F1-score of 0.9263, ranking 2nd out of 15 participating teams.

Although temporal topic modeling has been widely applied to scientific and legal texts, literary corpora have largely been overlooked in this regard. To address this issue, we analyze topic evolution in a subset of the Project Gutenberg (PG) corpus. We model this subset as a sequence of topic networks that capture the emergence, persistence, and interaction of thematic structures over decades. Using supervised topic representations, we predict nodes (topics) and edges (topic pairings) to forecast future topics and their co-occurrence. Our experiments demonstrate moderate to strong temporal persistence in topic connectivity patterns across three topic systems, with ROC-AUC and AP values consistently above 0.85. We find that the temporal span of topic networks significantly impacts predictive performance: longer spans improve the stability and recall of topic presence, while shorter spans better capture evolving topic relationships. Overall, our findings demonstrate the predictability of topics in literary texts over time.

Large Language Models often achieve strong performance by generating long intermediate chain-of-thought (CoT) reasoning. However, it remains unclear when a model’s final answer is actually determined during generation. If the answer is already fixed at an intermediate stage, subsequent reasoning tokens may constitute post-decision explanation, increasing inference cost and latency without improving correctness. We study the evolution of predicted answers over reasoning steps using *forced answer completion*, which elicits a model’s intermediate prediction at partial reasoning prefixes. Across a range of models and datasets, we find that the final answer is fixed from the earliest stages of generation in most cases, with **only 32% of answers changing after CoT**. Moreover, once the final answer switch does occur, models continue to generate substantial reasoning, producing **760 additional tokens per query** on average and accounting for around **40% of the total reasoning**. Motivated by these findings, we explore early-stopping strategies that halt generation once the answer has stabilized. We show that simple probe-based stopping can reduce reasoning token usage by **500 tokens per query** while incurring only a **2% drop in accuracy**. Together, our results indicate that a large portion of chain-of-thought generation functions as **post hoc explanation** and can be reduced with minimal impact on performance.

We present TTLab’s submission to the AlexandriaX-2026 Subtask 3 on Arabic MT error span detection and classification. Our system frames the task as token-level classification over surface forms, preserving character offsets to ensure exact alignment with the evaluation metric. To handle severe label imbalance, we employ a focal loss with class weighting and dialect-specific decoding thresholds. Among six Arabic pre-trained encoders, MARBERTv2 achieves the best overall performance of 40.8 and 40.91 on the development and test set, respectively, ranking 3rd out of all participating teams. While our system localizes error spans effectively, classification of rare error types remains challenging, highlighting the need for data augmentation for tail categories.

Arabic-language stance detection remains challenging, and previous shared-task systems have largely relied on multitask learning and ensembles. While these systems achieve state-of-the-art performance, their applicability and transferability are limited by the additional complexity introduced by multitask learning. To reduce this complexity, we introduce CLASP-Ar, which reformulates the task as cloze-style masked language modeling. In this approach, the target, predicted sentiment, and text are combined into a single prompt whose [MASK] prediction is restricted to a verbalizer-constrained label vocabulary. When evaluated on the StanceEval-2026: Arabic Stance Detection Shared Task, our proposed approach achieves an F_avg2 score of 71.36% on Track 1 and 74.14% on Track 2 test sets.

Argument Mining (AM) is a critical NLP task that remains significantly under-resourced in Arabic. This paper presents STAR-Ar, a BERT-BiLSTM-CRF architecture for argument discourse detection and classification, as our system for Daleel 2026, the inaugural Arabic argument mining shared task. The task requires the identification and classification of argumentative discourse units (ADUs) in debate and editorial texts. We jointly model these two objectives as a token-level sequence labeling task using a BERT-BiLSTM-CRF architecture that combines contextual transformer embeddings with structural transition constraints to support accurate span detection. STAR-Ar achieves an F1-score of 72.69 on validation and 73.7 on test data. Our domain-specific analysis shows that models trained exclusively on editorials underperform those trained on debates, a disparity we primarily attribute to the smaller size of the editorial dataset.

This paper provides an overview of Task 2 from the Social Media Mining for Health 2024 shared task (#SMM4H 2024), which focused on Named Entity Recognition (NER, Subtask 2a) and the joint task of NER and Relation Extraction (RE, Subtask 2b) for detecting adverse drug reactions (ADRs) in German, Japanese, and French texts written by patients. Participants were challenged with a few-shot learning scenario, necessitating models that can effectively generalize from limited annotated examples. Despite the diverse strategies employed by the participants, the overall performance across submissions from three teams highlighted significant challenges. The results underscored the complexity of extracting entities and relations in multi-lingual contexts, especially from the noisy and informal nature of user-generated content. Further research is required to develop robust systems capable of accurately identifying and associating ADR-related information in low-resource and multilingual settings.

The NLI4CT task at SemEval-2024 emphasizes the development of robust models for Natural Language Inference on Clinical Trial Reports (CTRs) using large language models (LLMs). This edition introduces interventions specifically targeting the numerical, vocabulary, and semantic aspects of CTRs. Our proposed system harnesses the capabilities of the state-of-the-art Mistral model (Jiang et al., 2023), complemented by an auxiliary model, to focus on the intricate input space of the NLI4CT dataset. Through the incorporation of numerical and acronym-based perturbations to the data, we train a robust system capable of handling both semantic-altering and numerical contradiction interventions. Our analysis on the dataset sheds light on the challenging sections of the CTRs for reasoning.

Scholarly Argumentation Mining (SAM) has recently gained attention due to its potential to help scholars with the rapid growth of published scientific literature. It comprises two subtasks: argumentative discourse unit recognition (ADUR) and argumentative relation extraction (ARE), both of which are challenging since they require e.g. the integration of domain knowledge, the detection of implicit statements, and the disambiguation of argument structure. While previous work focused on dataset construction and baseline methods for specific document sections, such as abstract or results, full-text scholarly argumentation mining has seen little progress. In this work, we introduce a sequential pipeline model combining ADUR and ARE for full-text SAM, and provide a first analysis of the performance of pretrained language models (PLMs) on both subtasks. We establish a new SotA for ADUR on the Sci-Arg corpus, outperforming the previous best reported result by a large margin (+7% F1). We also present the first results for ARE, and thus for the full AM pipeline, on this benchmark dataset. Our detailed error analysis reveals that non-contiguous ADUs as well as the interpretation of discourse connectors pose major challenges and that data annotation needs to be more consistent.