We introduce ***VLURes***, a multilingual benchmark for evaluating Vision-Language Models (VLMs) under *long-text grounding*: selecting and reasoning over the image-relevant subset of article-length text that contains distractors and ungrounded claims. *VLURes* contains **4,000** web-curated *image + long-text* pairs across **English (En), Japanese (Ja), Swahili (Sw), and Urdu (Ur)** and **10** topical categories, and defines **eight** tasks spanning image-only perception (OR, SU, RU, SS, IC) and image+text grounding (ITM, *Unrelatedness*, VQA). To construct web-realistic pairs, we apply language-adapted CLIP alignment to select representative images and filter weakly grounded pages. Across **10** proprietary and open VLMs evaluated under zero-shot and one-shot prompting, with and without rationales, the best model (GPT-4o) reaches **90.8%** overall accuracy but remains **6.7** points below human performance (**97.5%**) on Object Recognition, and cross-lingual sensitivity persists, while open models are substantially weaker and often lack reliable multilingual VL support. *VLURes* provides a practical testbed for long-text grounding and multilingual robustness in web-realistic agent settings.

We present MHRoBERT (Multistream HEAT over Recurrence over BERT), a hierarchical transformer architecture for longitudinal mental health monitoring that models self- and mutual excitation patterns in linguistic and temporal data across multivariate event streams relating to an individual’s mental health. To supply the model with complementary perspectives on each post, we apply a Large Language Model (LLM) based annotation to extract three streams from social media posts: emotional states, personal life events, and mental health symptoms. A central finding is that multi-task learning with these automatically-generated stream labels provides substantial, consistent improvements across all model architectures evaluated. Multistream information further consistently benefits simpler models not explicitly designed to exploit it: LLM baselines incorporating stream annotations improve macro F1 by 12.6% over text-only prompting. These results have direct implications for the CLPsych Shared Task on Moments of Change detection: multistream auxiliary supervision yields consistent, substantial gains regardless of architecture, suggesting it is a simple and portable strategy that future systems can readily adopt with minimal architectural changes. MHRoBERT additionally produces interpretable learned parameters across streams, revealing temporal interaction patterns between mental health indicators.

We provide an overview of the CLPsych 2026 Shared Task, which focuses on capturing and characterizing mental health dynamics from social media timelines through structured modeling of self-states. This year advances the longitudinal paradigm set by prior CLPsych shared tasks (2022, 2025), by integrating fine-grained psychological representation using the MIND framework. The task is organized into three main components: (1) post-level identification of adaptive and maladaptive self-states through ྀི elements and sub-elements, along with estimation of their presence; (2) timeline-level detection of Moments of Change, including both abrupt switches and gradual escalations based on ABCd element and sub-element combinations; and (3) sequence-level modeling, involving summarization of change processes over time and identification of recurrent dynamic signatures.

The rise of large language models (LLMs) generating human-like text has raised concerns about misuse, especially in low-resource languages like Urdu. To address this gap, we introduce the HLU dataset, which consists of three datasets: Document, Paragraph, and Sentence level. The document-level dataset contains 1,014 instances of human-written and LLM-generated articles across 13 domains, while the paragraph and sentence-level datasets each contain 667 instances. We conducted both human and automatic evaluations. In the human evaluation, the average accuracy at the document level was 35%, while at the paragraph and sentence levels, accuracies were 75.68% and 88.45%, respectively. For automatic evaluation, we finetuned the XLMRoBERTa model for both monolingual and multilingual settings achieving consistent results in both. Additionally, we assessed the performance of GPT4 and Claude3Opus using zero-shot prompting. Our experiments and evaluations indicate that distinguishing between human and machine-generated text is challenging for both humans and LLMs, marking a significant step in addressing this issue in Urdu.

We provide an overview of the CLPsych 2025 Shared Task, which focuses on capturing mental health dynamics from social media timelines. Building on CLPsych 2022’s longitudinal modeling approach, this work combines monitoring mental states with evidence and summary generation through four subtasks: (A.1) Evidence Extraction, highlighting text spans reflecting adaptive or maladaptive self-states; (A.2) Well-Being Score Prediction, assigning posts a 1 to 10 score based on social, occupational, and psychological functioning; (B) Post-level Summarization of the interplay between adaptive and maladaptive states within individual posts; and (C) Timeline-level Summarization capturing temporal dynamics of self-states over posts in a timeline. We describe key findings and future directions.

Paraphrase detection is a task to identify if two sentences are semantically similar or not. It plays an important role in maintaining the integrity of written work such as plagiarism detection and text reuse detection. Formerly, researchers focused on developing large corpora for English. However, no research has been conducted on sentence-level paraphrase detection in low-resource Pashto language. To bridge this gap, we introduce the first fully manually annotated Pashto sentential paraphrase detection corpus collected from authentic cases in journalism covering 10 different domains, including Sports, Health, Environment, and more. Our proposed corpus contains 6,727 sentences, encompassing 3,687 paraphrased and 3,040 non-paraphrased. Experimental findings reveal that our proposed corpus is sufficient to train XLM-RoBERTa to accurately detect paraphrased sentence pairs in Pashto with an F1 score of 84%. To compare our corpus with those in other languages, we also applied our fine-tuned model to the Indonesian and English paraphrase datasets in a zero-shot manner, achieving F1 scores of 82% and 78%, respectively. This result indicates that the quality of our corpus is not less than commonly used datasets. It‘s a pioneering contribution to the field. We will publicize a subset of 1,800 instances from our corpus, free from any licensing issues.