Tobias Lorenz

The Oracle's Gambit casts responsible AI release as a bilevel Stackelberg game, finding that for dual-use models the decisive lever is the sequencing of access — pre-releasing to defenders to open a protective capability gap — rather than the deploy-or-withhold threshold.

Certified Circuits identify the parts of a neural network that truly represent a concept, producing explanations that are more reliable, compact, and robust to changes in the data used to discover them.

LLM panels running the Delphi protocol achieve strong calibration (r=0.87–0.95) against benchmark ground truth and align closely with human expert panels, reducing elicitation time from months to minutes.

MIBP-Cert uses mixed-integer bilinear programming to compute sound, deterministic robustness bounds during training, handling complex threat models including discrete and continuous data perturbations.

3DCertify, the first verifier for point cloud models, certifies robustness against a wide range of semantic 3D transformations for both classification and part segmentation.

A GAN that synthesizes CNN features conditioned on class-level semantic information, enabling effective generalized zero-shot learning without labeled examples of unseen classes.