Giuseppe Bruno

activity

20242026

collaborators

8 papers

math.AP2026

Morrey's problem in $R^{2 \times 4}$ and $R_{sym}^{3 \times 3}$

Andrea Agazzi, Giuseppe Bruno, André Guerra +1

We find an explicit rank-one convex non-quasiconvex integrand in $R^{2 \times 4}$ : to falsify the quasiconvexity inequality, we exhibit a map $T^{4} \to R^{2}$ …

math.AP2026

Quasiconvexity for the Dacorogna--Marcellini Energy

Giuseppe Bruno, Federico Pasqualotto

We prove that the planar Dacorogna--Marcellini energy $f_{γ} (A) = ∣ A ∣^{4} - 2 γ ∣ A ∣^{2} det A$ is quasiconvex exactly when it is rank-one convex, i.e. if and only if $∣ γ ∣ \leq \frac{2}{3}$ .…

cs.LG2026

Scaling Limits of Long-Context Transformers

Giuseppe Bruno, Shi Chen, Zhengjiang Lin +2

We study the long-context limit of softmax self-attention with a fixed query and a random context of $n$ i.i.d. keys on the sphere, viewing the inverse temperature $I^_{n}^{2}$ as the sc…

math.PR2026

Stochastic Scaling Limits and Synchronization by Noise in Deep Transformer Models

Andrea Agazzi, Giuseppe Bruno, Eloy Mosig GarcÃa +2

We prove pathwise convergence of the layerwise evolution of tokens in a finite-depth, finite-width transformer model with MultiLayer Perceptron (MLP) blocks to a continuous-time st…

cs.LG2025

A multiscale analysis of mean-field transformers in the moderate interaction regime

Giuseppe Bruno, Federico Pasqualotto, Andrea Agazzi

In this paper, we study the evolution of tokens through the depth of encoder-only transformer models at inference time by modeling them as a system of particles interacting in a me…