Philipp Alexander Kreer

collaborators

3 papers

cs.LG2026

Bayesian Influence Functions for Hessian-Free Data Attribution

Philipp Alexander Kreer, Wilson Wu, Maxwell Adam +2

Classical influence functions face significant challenges when applied to deep neural networks, primarily due to non-invertible Hessians and high-dimensional parameter spaces. We p…

cs.AI2025

Noise Injection Reveals Hidden Capabilities of Sandbagging Language Models

Cameron Tice, Philipp Alexander Kreer, Nathan Helm-Burger +7

Capability evaluations play a crucial role in assessing and regulating frontier AI systems. The effectiveness of these evaluations faces a significant challenge: strategic underper…

hep-ph2025

One-Loop QCD Corrections to $u ˉ d \to t t ˉ W$ at $O (ε^{2})$

Matteo Becchetti, Maximilian Delto, Sara Ditsch +3

We present a computation of the one-loop QCD corrections to top-quark pair production in association with a $W$ boson, including terms up to order $ε^{2}$ in dimensional re…

Bayesian Influence Functions for Hessian-Free Data Attribution

Noise Injection Reveals Hidden Capabilities of Sandbagging Language Models

One-Loop QCD Corrections to uˉd→ttˉW at O(ε2)

One-Loop QCD Corrections to $u ˉ d \to t t ˉ W$ at $O (ε^{2})$