11 papers
Modelpedia: A Catalog of Model Findings for the Meta-Science of AI
Franciszek Bernat, Dawid Płudowski, Michał Jan Włodarczyk +6
Scientific knowledge about AI models is produced faster than the community can organize it. Every few months a new foundation model reshapes the field and hundreds of papers, blogs…
Your CLIP has 164 dimensions of noise: Exploring the embeddings covariance eigenspectrum of contrastively pretrained vision-language transformers
Jakub Grzywaczewski, Dawid Płudowski, Przemysław Biecek
Contrastively pre-trained Vision-Language Models (VLMs) serve as powerful feature extractors. Yet, their shared latent spaces are prone to structural anomalies and act as repositor…
SwordBench: Evaluating Orthogonality of Steering Image Representations
Vladimir Zaigrajew, Dawid Pludowski, Hubert Baniecki +1
Steering or intervening on model representations at inference time to correct predictions is essential for AI interpretability and safety, yet existing evaluation protocols are lim…
Local Intrinsic Dimension Unveils Hallucinations in Diffusion Models
Bartlomiej Sobieski, Matthew Tivnan, Dawid Płudowski +4
Diffusion models are prone to generating structural hallucinations - samples that match the statistical properties of the training data yet defy underlying structural rules, result…
Trojan horse hunt in deep forecasting models: Insights from the European Space Agency competition
Krzysztof Kotowski, Ramez Shendy, Jakub Nalepa +10
Forecasting plays a crucial role in modern safety-critical applications, such as space operations. However, the increasing use of deep forecasting models introduces a new security…
Fake or Real: The Impostor Hunt in Texts for Space Operations
Agata Kaczmarek, Dawid Płudowski, Piotr Wilczyński +6
The "Fake or Real" competition hosted on Kaggle (https://www.kaggle.com/competitions/fake-or-real-the-impostor-hunt ) is the second part of a series of follow-up competitions and h…
Divide, Specialize, and Route: A New Approach to Efficient Ensemble Learning
Jakub Piwko, Jędrzej Ruciński, Dawid Płudowski +5
Ensemble learning has proven effective in boosting predictive performance, but traditional methods such as bagging, boosting, and dynamic ensemble selection (DES) suffer from high…
Trojan Horse Hunt in Time Series Forecasting for Space Operations
Krzysztof Kotowski, Ramez Shendy, Jakub Nalepa +6
This competition hosted on Kaggle (https://www.kaggle.com/competitions/trojan-horse-hunt-in-space) is the first part of a series of follow-up competitions and hackathons related to…
MASCOTS: Model-Agnostic Symbolic COunterfactual explanations for Time Series
Dawid Płudowski, Francesco Spinnato, Piotr Wilczyński +4
Counterfactual explanations provide an intuitive way to understand model decisions by identifying minimal changes required to alter an outcome. However, applying counterfactual met…
The Dark Patterns of Personalized Persuasion in Large Language Models: Exposing Persuasive Linguistic Features for Big Five Personality Traits in LLMs Responses
Wiktoria Mieleszczenko-Kowszewicz, Dawid Płudowski, Filip Kołodziejczyk +3
This study explores how the Large Language Models (LLMs) adjust linguistic features to create personalized persuasive outputs. While research showed that LLMs personalize outputs,…
Rethinking of Encoder-based Warm-start Methods in Hyperparameter Optimization
Dawid Płudowski, Antoni Zajko, Anna Kozak +1
Effectively representing heterogeneous tabular datasets for meta-learning purposes remains an open problem. Previous approaches rely on predefined meta-features, for example, stati…