2 citations · 5 across the 15 of their papers we have counts for
15 papers
Timestep-Conditioned Transformers for Global Weather Forecasting
Sam Levang, Fran Bartolic, Ty Dickinson +3
Existing machine-learning weather forecasting models rely on predetermined and fixed autoregressive timesteps. The choice of model timestep involves a fundamental trade-off: shorte…
Demonstrating Generalization Failures via Mixtures of Conditional Policies
Jou Barzdukas, Jack Peck, Julian Schulz +3
Post-training of frontier language models is conducted on curated task suites, and inevitably leaves a distribution shift between training and deployment environments. This exposes…
Recursive Scaling in Masked Diffusion Models
Alba Carballo-Castro, Julianna Piskorz, Paulius Rauba +2
Masked diffusion models (MDMs) generate sequences by iteratively refining a partially masked state and committing tokens in parallel. We introduce recursion in MDMs and propose new…
Tiny Autoregressive Recursive Models
Paulius Rauba, Claudio Fanconi, Mihaela van der Schaar
Tiny Recursive Models (TRMs) have recently demonstrated remarkable performance on ARC-AGI, showing that very small models can compete against large foundation models through a two-…
No More, No Less: Least-Privilege Language Models
Paulius Rauba, Dominykas Seputis, Patrikas Vanagas +1
Least privilege is a core security principle: grant each request only the minimum access needed to achieve its goal. Deployed language models almost never follow it, instead being…
Multi-Agent Systems Should be Treated as Principal-Agent Problems
Paulius Rauba, Simonas Cepenas, Mihaela van der Schaar
Consider a multi-agent systems setup in which a principal (a supervisor agent) assigns subtasks to specialized agents and aggregates their responses into a single system-level outp…
Probabilistic Transformers for Joint Modeling of Global Weather Dynamics and Decision-Centric Variables
Paulius Rauba, Viktor Cikojevic, Fran Bartolic +3
Weather forecasts sit upstream of high-stakes decisions in domains such as grid operations, aviation, agriculture, and emergency response. Yet forecast users often face a difficult…
Visualizing token importance for black-box language models
Paulius Rauba, Qiyao Wei, Mihaela van der Schaar
We consider the problem of auditing black-box large language models (LLMs) to ensure they behave reliably when deployed in production settings, particularly in high-stakes domains…
Deep Hierarchical Learning with Nested Subspace Networks for Large Language Models
Paulius Rauba, Mihaela van der Schaar
Large neural networks are typically trained for a fixed computational budget, creating a rigid trade-off between performance and efficiency that is ill-suited for deployment in res…
Statistical Hypothesis Testing for Auditing Robustness in Language Models
Paulius Rauba, Qiyao Wei, Mihaela van der Schaar
Consider the problem of testing whether the outputs of a large language model (LLM) system change under an arbitrary intervention, such as an input perturbation or changing the mod…
Quantifying perturbation impacts for large language models
Paulius Rauba, Qiyao Wei, Mihaela van der Schaar
We consider the problem of quantifying how an input perturbation impacts the outputs of large language models (LLMs), a fundamental task for model reliability and post-hoc interpre…
Self-Healing Machine Learning: A Framework for Autonomous Adaptation in Real-World Environments
Paulius Rauba, Nabeel Seedat, Krzysztof Kacprzyk +1
Real-world machine learning systems often encounter model performance degradation due to distributional shifts in the underlying data generating process (DGP). Existing approaches…
Context-Aware Testing: A New Paradigm for Model Testing with Large Language Models
Paulius Rauba, Nabeel Seedat, Max Ruiz Luyten +1
The predominant de facto paradigm of testing ML models relies on either using only held-out data to compute aggregate evaluation metrics or by assessing the performance on differen…
Redefining Digital Health Interfaces with Large Language Models
Fergus Imrie, Paulius Rauba, Mihaela van der Schaar
Digital health tools have the potential to significantly improve the delivery of healthcare services. However, their adoption remains comparatively limited due, in part, to challen…
Memes in the Wild: Assessing the Generalizability of the Hateful Memes Challenge Dataset
Hannah Rose Kirk, Yennie Jun, Paulius Rauba +7
Hateful memes pose a unique challenge for current machine learning systems because their message is derived from both text- and visual-modalities. To this effect, Facebook released…