← Search

Vincent Létourneau

5 accepted papers

2026

The Geometry and Topology of Circuits: the Manifolds of Modular Addition

ICLR 2026poster

The Clock and Pizza interpretations, associated with architectures differing in either uniform or learnable attention, were introduced to argue that different architectural designs can yield distinct circuits for modular addition. In this work, we show that this is not the case, and that both the un…

Cited by 0SourceScholar
2025

Uncovering a Universal Abstract Algorithm for Modular Addition in Neural Networks

NeurIPS 2025poster

We propose a testable universality hypothesis, asserting that seemingly disparate neural network solutions observed in the simple task of modular addition actually reflect a common abstract algorithm. While prior work interpreted variations in neuron-level representations as evidence for distinct al…

Cited by 0SourceScholar
2024

Graph Positional and Structural Encoder

ICML 2024poster

Positional and structural encodings (PSE) enable better identifiability of nodes within a graph, rendering them essential tools for empowering modern GNNs, and in particular graph Transformers. However, designing PSEs that work optimally for all graph prediction tasks is a challenging and unsolved p…

2021

Directional Graph Networks

ICML 2021oral

The lack of anisotropic kernels in graph neural networks (GNNs) strongly limits their expressiveness, contributing to well-known issues such as over-smoothing. To overcome this limitation, we propose the first globally consistent anisotropic kernels for GNNs, allowing for graph convolutions that are…

2021

Rethinking Graph Transformers with Spectral Attention

NeurIPS 2021poster

In recent years, the Transformer architecture has proven to be very successful in sequence processing, but its application to other data structures, such as graphs, has remained limited due to the difficulty of properly defining positions. Here, we present the \textit{Spectral Attention Network} (S…