5 papers
Deep Learning as Neural Low-Degree Filtering: A Spectral Theory of Hierarchical Feature Learning
Yatin Dandi, Matteo Vilucchio, Luca Arnaboldi +2
Understanding how deep neural networks learn useful internal representations from data remains a central open problem in the theory of deep learning. We introduce Neural Low-Degree…
Beyond Red-Teaming: Formal Guarantees of LLM Guardrail Classifiers
Nikita Kezins, Urbas Ekka, Pascal Berrang +1
Guardrail Classifiers defend production language models against harmful behavior, but although results seem promising in testing, they provide no formal guarantees. Providing forma…
ColBERT-Zero: To Pre-train Or Not To Pre-train ColBERT models
Antoine Chaffin, Luca Arnaboldi, Amélie Chatelain +1
Current state-of-the-art multi-vector models are obtained through a small Knowledge Distillation (KD) training step on top of strong single-vector models, leveraging the large-scal…
Asymptotics of SGD in Sequence-Single Index Models and Single-Layer Attention Networks
Luca Arnaboldi, Bruno Loureiro, Ludovic Stephan +2
We study the dynamics of stochastic gradient descent (SGD) for a class of sequence models termed Sequence Single-Index (SSI) models, where the target depends on a single direction…
Repetita Iuvant: Data Repetition Allows SGD to Learn High-Dimensional Multi-Index Functions
Luca Arnaboldi, Yatin Dandi, Florent Krzakala +2
Neural networks can identify low-dimensional relevant structures within high-dimensional noisy data, yet our mathematical understanding of how they do so remains scarce. Here, we i…