Mihai Gabriel Constantin

The automatic detection of persuasive strategies in advertisements presents a uniquely multimodal challenge at the intersection of vision, language, and social cognition. While recent advances in Large Vision-Language Models (LVLMs) offer promising capabilities for such tasks, current approaches often rely on restrictive evaluation schemes that do not reflect the inherently multi-label nature of persuasive messaging. In this work, we reframe persuasion strategy detection as a genuine multi-label classification problem and propose a principled evaluation framework to enhance interpretability and robustness. We apply this approach to both image and video datasets, examine their characteristics, and introduce novel input-agnostic baselines that achieve macro F1 scores of 0.082 and 0.289 on the respective test sets. As part of this analysis, we study label co-occurrence patterns and dataset ambiguities, providing insights that inform both model interpretation and future dataset design. To assess the native capabilities of LVLMs, we benchmark three open source models—PaliGemma, PaliGemma2, and Qwen2.5-VL—on the image persuasion dataset. Under zero-shot conditions, we demonstrate that querying each strategy individually with a logit-based decision threshold outperforms guided text generation. The best-performing zero-shot model, Qwen2.5-VL, achieves a macro F1 score of 0.227 and a sample F1 score of 0.234 on the image test set, and 0.381 and 0.400 respectively on the video test set. We further explore and compare two lightweight fine-tuning strategies that update only small subsets of model parameters while keeping the remaining weights frozen: fine-tuning of the image-to-text tokens linear projection and Low Rank Adaptation (LoRA) of the language model. Linear projector fine-tuning yields a top macro F1 score of 0.396 on the image test set, marking a substantial improvement over zero-shot performance. To evaluate cross-modal generalization, we apply fine-tuned image models to the video dataset. Our experiments reveal that, while projection-based fine-tuning enables partial knowledge transfer from image to video (macro F1 = 0.416 on a testing subset), LoRA adaptation severely disrupts cross-modal performance (macro F1 = 0.189). Finally, we perform a per-strategy performance analysis, looking into annotator- and data-centric factors that may influence LVLM performance. These findings highlight the viability of open-weight LVLMs for fine-grained persuasion analysis and suggest efficient pathways for domain-specific adaptation under realistic resource constraints.

One of the powers of visual media lies in its ability to create a lasting impression on the viewer’s memory. In this digital age, where media is abundant and attention spans are fleeting, the task of predicting which content will stick in the viewer’s mind has become a critical challenge in computer vision. Computational memorability seeks to address this by developing models that estimate how memorable a piece of media is likely to be. In this review we focus on the MediaEval Predicting Video Memorability benchmark, a recurring evaluation task that has run annually since 2018. This benchmark provides a unique and consistent framework for researchers to compare and refine their memorability prediction techniques using standardised datasets and metrics. Its reproducible framework has proven invaluable for tracking progress and fostering innovation in this rapidly evolving domain. We analyse the evolution of the benchmark across its 2018–2023 editions, discussing the challenges that still remain, such as the need for more interpretability in models and the difficulty of predicting subjective and context-dependent memorability. By analysing and synthesising the collective insights gained from this task, we endeavour to inspire new avenues of inquiry and drive progress towards a more comprehensive understanding of this topic.

The advent of generative networks and their adoption in numerous domains and communities have led to a wave of innovation and breakthroughs in AI and machine learning. Generative Adversarial Networks (GANs) have expanded the scope of what is possible with machine learning, allowing for new applications in areas such as computer vision, natural language processing, and creative AI. GANs, in particular, have been used for a wide range of tasks, including image and video generation, data augmentation, style transfer, and anomaly detection. They have also been used for medical imaging and drug discovery, where they can generate synthetic data to augment small datasets, reduce the need for expensive experiments, and lower the number of real patients that must be included in medical trials. Given these developments, we propose using the power of GANs to create and augment flow-based network traffic datasets. We evaluate a series of GAN architectures, including Wasserstein, conditional, energy-based, gradient penalty, and LSTM-GANs. We evaluate their performance on a set of flow-based network traffic data collected from 16 subjects who used their computers for home, work, and study purposes. The performance of these GAN architectures is described according to metrics that involve networking principles, data distribution among a collection of flows, and temporal data distribution. Given the tendency of network intrusion detection datasets to have a very imbalanced data distribution, i.e., a large number of samples in the “normal traffic” category and a comparatively low number of samples assigned to the “intrusion” categories, we test our GANs by augmenting the intrusion data and checking whether this helps intrusion detection neural networks in their task. We publish the resulting UPBFlow dataset and code on GitHub. ¹