Influenza A viruses (IAVs) evolve antigenically at a pace that requires frequent vaccine updates, yet the haemagglutination inhibition (HI) assays used to quantify antigenicity are labor-intensive and unscalable. As a result, genomic data vastly outpace available phenotypic labels, limiting the effectiveness of traditional supervised models. We hypothesize that combining pre-trained Protein Language Models (PLMs) with Semi-Supervised Learning (SSL) can retain high predictive accuracy even when labeled data are scarce. We evaluated two SSL strategies, Self-training and Label Spreading, against fully supervised baselines using four PLM-derived embeddings (ESM-2, ProtVec, ProtT5, ProtBert) applied to haemagglutinin (HA) sequences. A nested cross-validation framework simulated low-label regimes (25%, 50%, 75%, and 100% label availability) across four IAV subtypes (H1N1, H3N2, H5N1, H9N2). SSL consistently improved performance under label scarcity. Self-training with ProtVec produced the largest relative gains, showing that SSL can compensate for lower-resolution representations. ESM-2 remained highly robust, achieving F1 scores above 0.82 with only 25% labeled data, indicating that its embeddings capture key antigenic determinants. While H1N1 and H9N2 were predicted with high accuracy, the hypervariable H3N2 subtype remained challenging, although SSL mitigated the performance decline. These findings demonstrate that integrating PLMs with SSL can address the antigenicity labeling bottleneck and enable more effective use of unlabeled surveillance sequences, supporting rapid variant prioritization and timely vaccine strain selection.

Speech brain-computer interfaces (BCIs) aim to restore communication for people with paralysis by translating neural activity into text. Most systems use cascaded frameworks that decode phonemes before assembling sentences with an n-gram language model (LM), preventing joint optimization of all stages simultaneously. Here, we introduce an end-to-end Brain-to-Text (BIT) framework that translates neural activity into coherent sentences using a single differentiable neural network. Central to our approach is a cross-task, cross-species pretrained neural encoder, whose representations transfer to both attempted and imagined speech. In a cascaded setting with an n-gram LM, the pretrained encoder establishes a new state-of-the-art (SOTA) on the Brain-to-Text '24 and '25 benchmarks. Integrated end-to-end with audio large language models (LLMs) and trained with contrastive learning for cross-modal alignment, BIT reduces the word error rate (WER) of the prior end-to-end method from 24.69% to 10.22%. Notably, we find that small-scale audio LLMs markedly improve end-to-end decoding. Beyond record-setting performance, BIT aligns attempted and imagined speech embeddings to enable cross-task generalization. Altogether, our approach advances the integration of large, diverse neural datasets, paving the way for an end-to-end decoding framework that supports seamless, differentiable optimization.

As AI systems become more capable of complex agentic tasks, they also become more capable of pursuing undesirable objectives and causing harm. Previous work has attempted to catch these unsafe instances by interrogating models directly about their objectives and behaviors. However, the main weakness of trusting interrogations is that models can lie. We propose self-report fine-tuning (SRFT), a simple supervised fine-tuning technique that trains models to occasionally make factual mistakes, then admit them when asked. We show that the admission of factual errors in simple question-answering settings generalizes out-of-distribution (OOD) to the admission of hidden misaligned objectives in adversarial agentic settings. We evaluate SRFT in OOD stealth tasks, where models are instructed to complete a hidden misaligned objective alongside a user-specified objective without being caught by monitoring. After SRFT, models are more likely to confess the details of their hidden objectives when interrogated, even under strong pressure not to disclose them. Interrogation on SRFT models can detect hidden objectives with near-ceiling performance (F1 score = 0.98), while the baseline model lies when interrogated under the same conditions (F1 score = 0). Interrogation on SRFT models can further elicit the content of the hidden objective, recovering 28-100% details, compared to 0% details recovered in the baseline model and by prefilled assistant turn attacks. This provides a promising technique for promoting honesty propensity and incriminating misaligned AIs.

Social recommendation has been proven effective in addressing data sparsity in user-item interaction modeling by leveraging social networks. The recent integration of Graph Neural Networks (GNNs) has further enhanced prediction accuracy in contemporary social recommendation algorithms. However, many GNN-based approaches in social recommendation lack the ability to furnish meaningful explanations for their predictions. In this study, we confront this challenge by introducing SoREX, a self-explanatory GNN-based social recommendation framework. SoREX adopts a two-tower framework enhanced by friend recommendation, independently modeling social relations and user-item interactions, while jointly optimizing an auxiliary task to reinforce social signals. To offer explanations, we propose a novel ego-path extraction approach. This method involves transforming the ego-net of a target user into a collection of multi-hop ego-paths, from which we extract factor-specific and candidate-aware ego-path subsets as explanations. This process facilitates the summarization of detailed comparative explanations among different candidate items through intricate substructure analysis. Furthermore, we conduct explanation re-aggregation to explicitly correlate explanations with downstream predictions, imbuing our framework with inherent self-explainability. Comprehensive experiments conducted on four widely adopted benchmark datasets validate the effectiveness of SoREX in predictive accuracy. Additionally, qualitative and quantitative analyses confirm the efficacy of the extracted explanations in SoREX. Our code and data are available at https://github.com/antman9914/SoREX.

Internal clustering validity indices (ICVIs) assess clustering quality without ground truth labels. Comparative studies consistently find that no single ICVI outperforms others across datasets, leaving practitioners without principled ICVI selection. We argue that inconsistent ICVI performance arises because studies evaluate them based on matching human labels rather than measuring the quality of the discovered structure in the data, using datasets without formally quantifying the structure type and quality. Structure type refers to the mathematical organisation in data that clustering aims to discover. Validity theory requires a theoretical definition of clustering quality, which depends on structure type. We demonstrate this through the first validity assessment of clustering quality measures for correlation patterns, a structure type that arises from clustering time series by correlation relationships. We formalise 23 canonical correlation patterns as the theoretical optimal clustering and use synthetic data modelling this structure with controlled perturbations to evaluate validity across content, criterion, construct, and external validity. Our findings show that Silhouette Width Criterion (SWC) and Davies-Bouldin Index (DBI) are valid for correlation patterns, whilst Calinski-Harabasz (VRC) and Pakhira-Bandyopadhyay-Maulik (PBM) indices fail. Simple Lp norm distances achieve validity, whilst correlation-specific functions fail structural, criterion, and external validity. These results differ from previous studies where VRC and PBM performed well, demonstrating that validity depends on structure type. Our structure-type-specific validation method provides both practical guidance (quality thresholds SWC>0.9, DBI<0.15) and a methodological template for establishing validity for other structure types.

Long context reasoning in large language models (LLMs) has demonstrated enhancement of their cognitive capabilities via chain-of-thought (CoT) inference. Training such models is usually done via reinforcement learning with verifiable rewards (RLVR) in reasoning based problems, like math and programming. However, RLVR is limited by several bottlenecks, such as, lack of dense reward, and inadequate sample efficiency. As a result, it requires significant compute resources in post-training phase. To overcome these limitations, in this work, we propose \textbf{Semantic Soft Bootstrapping (SSB)}, a self-distillation technique, in which the same base language model plays the role of both teacher and student, but receives different semantic contexts about the correctness of its outcome at training time. The model is first prompted with a math problem and several rollouts are generated. From them, the correct and most common incorrect response are filtered, and then provided to the model in context to produce a more robust, step-by-step explanation with a verified final answer. This pipeline automatically curates a paired teacher-student training set from raw problem-answer data, without any human intervention. This generation process also produces a sequence of logits, which is what the student model tries to match in the training phase just from the bare question alone. In our experiment, Qwen2.5-3B-Instruct on GSM8K dataset via parameter-efficient fine-tuning. We then tested its accuracy on MATH500, and AIME2024 benchmarks. Our experiments show a jump of 10.6%, and 10% improvements in accuracy, respectively, over group relative policy optimization (GRPO), which is a commonly used RLVR algorithm. Our code is available at https://github.com/purbeshmitra/semantic-soft-bootstrapping, and the model, curated dataset is available at https://huggingface.co/purbeshmitra/semantic-soft-bootstrapping.

Large Language Models (LLMs) have become a key topic in AI and NLP, transforming sectors like healthcare, finance, education, and marketing by improving customer service, automating tasks, providing insights, improving diagnostics, and personalizing learning experiences. Information extraction from clinical records is a crucial task in digital healthcare. Although traditional NLP techniques have been used for this in the past, they often fall short due to the complexity, variability of clinical language, and high inner semantics in the free clinical text. Recently, Large Language Models (LLMs) have become a powerful tool for better understanding and generating human-like text, making them highly effective in this area. In this paper, we explore the ability of open-source multilingual LLMs to understand EHRs (Electronic Health Records) in Italian and help extract information from them in real-time. Our detailed experimental campaign on comorbidity extraction from EHR reveals that some LLMs struggle in zero-shot, on-premises settings, and others show significant variation in performance, struggling to generalize across various diseases when compared to native pattern matching and manual annotations.

The rapid progress of Large Language Models (LLMs) has transformed natural language processing and broadened its impact across research and society. Yet, systematic evaluation of these models, especially for languages beyond English, remains limited. "Challenging the Abilities of LAnguage Models in ITAlian" (CALAMITA) is a large-scale collaborative benchmarking initiative for Italian, coordinated under the Italian Association for Computational Linguistics. Unlike existing efforts that focus on leaderboards, CALAMITA foregrounds methodology: it federates more than 80 contributors from academia, industry, and the public sector to design, document, and evaluate a diverse collection of tasks, covering linguistic competence, commonsense reasoning, factual consistency, fairness, summarization, translation, and code generation. Through this process, we not only assembled a benchmark of over 20 tasks and almost 100 subtasks, but also established a centralized evaluation pipeline that supports heterogeneous datasets and metrics. We report results for four open-weight LLMs, highlighting systematic strengths and weaknesses across abilities, as well as challenges in task-specific evaluation. Beyond quantitative results, CALAMITA exposes methodological lessons: the necessity of fine-grained, task-representative metrics, the importance of harmonized pipelines, and the benefits and limitations of broad community engagement. CALAMITA is conceived as a rolling benchmark, enabling continuous integration of new tasks and models. This makes it both a resource -- the most comprehensive and diverse benchmark for Italian to date -- and a framework for sustainable, community-driven evaluation. We argue that this combination offers a blueprint for other languages and communities seeking inclusive and rigorous LLM evaluation practices.

The rise of supply chain attacks via malicious Python packages demands robust detection solutions. Current approaches, however, overlook two critical challenges: robustness against adversarial source code transformations and adaptability to the varying false positive rate (FPR) requirements of different actors, from repository maintainers (requiring low FPR) to enterprise security teams (higher FPR tolerance). We introduce a robust detector capable of seamless integration into both public repositories like PyPI and enterprise ecosystems. To ensure robustness, we propose a novel methodology for generating adversarial packages using fine-grained code obfuscation. Combining these with adversarial training (AT) enhances detector robustness by 2.5x. We comprehensively evaluate AT effectiveness by testing our detector against 122,398 packages collected daily from PyPI over 80 days, showing that AT needs careful application: it makes the detector more robust to obfuscations and allows finding 10% more obfuscated packages, but slightly decreases performance on non-obfuscated packages. We demonstrate production adaptability of our detector via two case studies: (i) one for PyPI maintainers (tuned at 0.1% FPR) and (ii) one for enterprise teams (tuned at 10% FPR). In the former, we analyze 91,949 packages collected from PyPI over 37 days, achieving a daily detection rate of 2.48 malicious packages with only 2.18 false positives. In the latter, we analyze 1,596 packages adopted by a multinational software company, obtaining only 1.24 false positives daily. These results show that our detector can be seamlessly integrated into both public repositories like PyPI and enterprise ecosystems, ensuring a very low time budget of a few minutes to review the false positives. Overall, we uncovered 346 malicious packages, now reported to the community.

This study investigates logistic regression, linear support vector machine, multinomial Naive Bayes, and Bernoulli Naive Bayes for classifying Libyan dialect utterances gathered from Twitter. The dataset used is the QADI corpus, which consists of 540,000 sentences across 18 Arabic dialects. Preprocessing challenges include handling inconsistent orthographic variations and non-standard spellings typical of the Libyan dialect. The chi-square analysis revealed that certain features, such as email mentions and emotion indicators, were not significantly associated with dialect classification and were thus excluded from further analysis. Two main experiments were conducted: (1) evaluating the significance of meta-features extracted from the corpus using the chi-square test and (2) assessing classifier performance using different word and character n-gram representations. The classification experiments showed that Multinomial Naive Bayes (MNB) achieved the highest accuracy of 85.89% and an F1-score of 0.85741 when using a (1,2) word n-gram and (1,5) character n-gram representation. In contrast, Logistic Regression and Linear SVM exhibited slightly lower performance, with maximum accuracies of 84.41% and 84.73%, respectively. Additional evaluation metrics, including log loss, Cohen kappa, and Matthew correlation coefficient, further supported the effectiveness of MNB in this task. The results indicate that carefully selected n-gram representations and classification models play a crucial role in improving the accuracy of Libyan dialect identification. This study provides empirical benchmarks and insights for future research in Arabic dialect NLP applications.