The primary goal of a news headline is to summarize an event in as few words as possible. Depending on the media outlet, a headline can serve as a means to objectively deliver a summary or improve its visibility. For the latter, specific publications may employ stylistic approaches that incorporate the use of sarcasm, irony, and exaggeration, key elements of a satirical approach. As such, even the headline must reflect the tone of the satirical main content. Current approaches for the Romanian language tend to detect the non-conventional tone (i.e., satire and clickbait) of the news content by combining both the main article and the headline. Because we consider a headline to be merely a brief summary of the main article, we investigate in this paper the presence of satirical tone in headlines alone, testing multiple baselines ranging from standard machine learning algorithms to deep learning models. Our experiments show that Bidirectional Transformer models outperform both standard machine-learning approaches and Large Language Models (LLMs), particularly when the meta-learning Reptile approach is employed.

Data Science and Engineering Lab Universit e libre de Bruxelles Brussels, Belgium gaspard.merten@ulb.be Data Science and Engineering Lab Universit e libre de Bruxelles Brussels, Belgium gilles.dejaegere@ulb.be Data Science and Engineering Lab Universit e libre de Bruxelles Brussels, Belgium mahmoud.sakr@ulb.be Abstract --Recent research in Large Language Models (LLMs), has had a profound impact across various fields, including mobility data science. This paper explores the and experiment with different approaches to using LLMs for analyzing AIS data. We propose a set of carefully designed queries to assess the reasoning capabilities of LLMs in this kind of tasks. Further, we experiment with four different methods: (1) using LLMs as a natural language interface to a spatial database, (2) reasoning on raw data, (3) reasoning on compressed trajectories, and (4) reasoning on semantic trajectories. We investigate the strengths and weaknesses for the four methods, and discuss the findings. The goal is to provide valuable insights for both researchers and practitioners on selecting the most appropriate LLM-based method depending on their specific data analysis objectives. The significant development in artificial machine learning has also opened the way to new approaches to solve real-world geospatial problems. In particular, Large Language Models (LLMs) have emerged as powerful tools for understanding and generating human-like text. These models have demonstrated remarkable abilities in natural language processing tasks, from answering complex queries to summarizing and interpreting information in various domains. This exponential increase of LLMs usage can also be witnessed in the domain of Geographic Information Systems (GIS) in recent years.

Machine unlearning is concerned with the task of removing knowledge learned from particular data points from a trained model. In the context of large language models (LLMs), unlearning has recently received increased attention, particularly for removing knowledge about named entities from models for privacy purposes. While various approaches have been proposed to address the unlearning problem, most existing approaches treat all data points to be unlearned equally, i.e., unlearning that Montreal is a city in Canada is treated exactly the same as unlearning the phone number of the first author of this paper. In this work, we show that this all data is equal assumption does not hold for LLM unlearning. We study how the success of unlearning depends on the frequency of the knowledge we want to unlearn in the pre-training data of a model and find that frequency strongly affects unlearning, i.e., more frequent knowledge is harder to unlearn. Additionally, we uncover a misalignment between probability and generation-based evaluations of unlearning and show that this problem worsens as models become larger. Overall, our experiments highlight the need for better evaluation practices and novel methods for LLM unlearning that take the training data of models into account.

Preference learning is critical for aligning large language models (LLMs) with human values, yet its success hinges on high-quality datasets comprising three core components: Preference \textbf{A}nnotations, \textbf{I}nstructions, and \textbf{R}esponse Pairs. Current approaches conflate these components, obscuring their individual impacts and hindering systematic optimization. In this work, we propose \textbf{AIR}, a component-wise analysis framework that systematically isolates and optimizes each component while evaluating their synergistic effects. Through rigorous experimentation, AIR reveals actionable principles: annotation simplicity (point-wise generative scoring), instruction inference stability (variance-based filtering across LLMs), and response pair quality (moderate margins + high absolute scores). When combined, these principles yield +5.3 average gains over baseline method, even with only 14k high-quality pairs. Our work shifts preference dataset design from ad hoc scaling to component-aware optimization, offering a blueprint for efficient, reproducible alignment.

Large Language Models (LLMs) inherently reflect the vast data distributions they encounter during their pre-training phase. As this data is predominantly sourced from the web, there is a high chance it will be skewed towards high-resourced languages and cultures, such as those of the West. Consequently, LLMs often exhibit a diminished understanding of certain communities, a gap that is particularly evident in their knowledge of Arabic and Islamic cultures. This issue becomes even more pronounced with increasingly under-represented topics. To address this critical challenge, we introduce PalmX 2025, the first shared task designed to benchmark the cultural competence of LLMs in these specific domains. The task is composed of two subtasks featuring multiple-choice questions (MCQs) in Modern Standard Arabic (MSA): General Arabic Culture and General Islamic Culture. These subtasks cover a wide range of topics, including traditions, food, history, religious practices, and language expressions from across 22 Arab countries. The initiative drew considerable interest, with 26 teams registering for Subtask 1 and 19 for Subtask 2, culminating in nine and six valid submissions, respectively. Our findings reveal that task-specific fine-tuning substantially boosts performance over baseline models. The top-performing systems achieved an accuracy of 72.15% on cultural questions and 84.22% on Islamic knowledge. Parameter-efficient fine-tuning emerged as the predominant and most effective approach among participants, while the utility of data augmentation was found to be domain-dependent.

Post-training of Large Language Models (LMs) often prioritizes accuracy and helpfulness at the expense of diversity. This creates a tension: while post-training improves response quality, it also sharpens output distributions and reduces the range of ideas, limiting the usefulness of LMs in creative and exploratory tasks such as brainstorming, storytelling, or problem solving. We address this challenge with Diversity-Aware Reinforcement Learning (DARLING), a framework that jointly optimizes for response quality and semantic diversity. At its core, DARLING introduces a learned partition function to measure diversity beyond surface-level lexical variations. This diversity signal is then combined with a quality reward during online reinforcement learning, encouraging models to generate outputs that are both high-quality and distinct. Experiments across multiple model families and sizes show that DARLING generalizes to two regimes: non-verifiable tasks (instruction following and creative writing) and verifiable tasks (competition math). On five benchmarks in the first setting, DARLING consistently outperforms quality-only RL baselines, producing outputs that are simultaneously of higher quality and novelty. In the second setting, DARLING achieves higher pass@1 (solution quality) and pass@k (solution variety). Most strikingly, explicitly optimizing for diversity catalyzes exploration in online RL, which manifests itself as higher-quality responses.

Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) have empowered large language models (LLMs) to tackle challenging reasoning tasks such as mathematics and programming. RLVR leverages verifiable outcome rewards to guide policy optimization, enabling LLMs to progressively improve output quality in a grounded and reliable manner. Despite its promise, the RLVR paradigm poses significant challenges, as existing methods often suffer from sparse reward signals and unstable policy gradient updates, particularly in RL-based approaches. To address the challenges, we propose $\textbf{PACS}$, a novel RLVR framework that achieves im$\textbf{P}$licit $\textbf{A}$ctor $\textbf{C}$ritic coupling via a $\textbf{S}$upervised learning framework. By treating the outcome reward as a predictable label, we reformulate the RLVR problem into a supervised learning task over a score function parameterized by the policy model and optimized using cross-entropy loss. A detailed gradient analysis shows that this supervised formulation inherently recovers the classical policy gradient update while implicitly coupling actor and critic roles, yielding more stable and efficient training. Benchmarking on challenging mathematical reasoning tasks, PACS outperforms strong RLVR baselines, such as PPO and GRPO, achieving superior reasoning performance. For instance, PACS achieves 59.78\% at pass@256 on AIME 2025, representing improvements of 13.32 and 14.36 points over PPO and GRPO. This simple yet powerful framework offers a promising avenue for LLMs post-training with verifiable rewards. Our code and data are available as open source at https://github.com/ritzz-ai/PACS.

Large Language and Vision Models using a Mixture-of-Experts (MoE) architecture pose significant challenges for deployment due to their computational and memory demands. Mixed Precision Quantization assigns different precisions to different layers of an LLM/VLM based on layer sensitivity and importance within the model. In this work, we propose a Post Training Quantization algorithm, MoPEQ, that assigns optimal bit width to each expert. Our method balances accuracy and model size by analyzing each expert's sensitivity using Hessian trace approximation instead of relying on the activation frequency of the expert. This per-expert granularity approach clusters similar experts to maintain model performance while reducing memory requirements. The experimental results on VLMEvalKit benchmark datasets using State-of-the-art VLMs Deepseek-VL2 -tiny, -small, -base, and MolmoE models demonstrate that our mixed precision quantized MoEs achieve competitive accuracy with substantial improvements in memory footprint compared to uniform-precision baseline methods. W e perform a comprehensive study to analyze the impact of expert activation frequency and sensitivity using Hessian trace approximation at both layer-wise and model-wide expert precision allocation of 2, 3, and 4 bits to provide a thorough understanding of mixed precision quantization of VLM-MoEs. The code is available here.

Semantic evaluation in low-resource languages remains a major challenge in NLP. While sentence transformers have shown strong performance in high-resource settings, their effectiveness in Indic languages is underexplored due to a lack of high-quality benchmarks. To bridge this gap, we introduce L3Cube-IndicHeadline-ID, a curated headline identification dataset spanning ten low-resource Indic languages: Marathi, Hindi, Tamil, Gujarati, Odia, Kannada, Malayalam, Punjabi, Telugu, Bengali and English. Each language includes 20,000 news articles paired with four headline variants: the original, a semantically similar version, a lexically similar version, and an unrelated one, designed to test fine-grained semantic understanding. The task requires selecting the correct headline from the options using article-headline similarity. We benchmark several sentence transformers, including multilingual and language-specific models, using cosine similarity. Results show that multilingual models consistently perform well, while language-specific models vary in effectiveness. Given the rising use of similarity models in Retrieval-Augmented Generation (RAG) pipelines, this dataset also serves as a valuable resource for evaluating and improving semantic understanding in such applications. Additionally, the dataset can be repurposed for multiple-choice question answering, headline classification, or other task-specific evaluations of LLMs, making it a versatile benchmark for Indic NLP. The dataset is shared publicly at https://github.com/l3cube-pune/indic-nlp

The complexity of traditional power system analysis workflows presents significant barriers to efficient decision-making in modern electric grids. This paper presents GridMind, a multi-agent AI system that integrates Large Language Models (LLMs) with deterministic engineering solvers to enable conversational scientific computing for power system analysis. The system employs specialized agents coordinating AC Optimal Power Flow and N-1 contingency analysis through natural language interfaces while maintaining numerical precision via function calls. GridMind addresses workflow integration, knowledge accessibility, context preservation, and expert decision-support augmentation. Experimental evaluation on IEEE test cases demonstrates that the proposed agentic framework consistently delivers correct solutions across all tested language models, with smaller LLMs achieving comparable analytical accuracy with reduced computational latency. This work establishes agentic AI as a viable paradigm for scientific computing, demonstrating how conversational interfaces can enhance accessibility while preserving numerical rigor essential for critical engineering applications.