Great news! Deutsche Bahn will build an additional rail line for the 151 kilometers from Hannover to Hamburg. It will be finished by...wait for it...2050!

Wir setzen mit der Einkommensteuer klare Schwerpunkte: Familien und Menschen mit kleinen und mittleren Einkommen werden damit mehr Geld in der Tasche haben. Um das zu finanzieren, wird die Reichensteuer künftig früher greifen und eine Superreichensteuer ab 280.000 € eingeführt.

NEW paper at #Neurips 2026! We cloned the voices of non-native English speakers (n=86) with ElevenLabs, Coqui-XTTS and ChatterBox. In 4,000 human ratings, the clones sounded warmer, more authoritative, and even more *humanlike* than the real speakers. Listeners also said they would trust the clones more.

Backprop has been the only credit assignment algorithm capable of training large neural nets. Introducing Dust, the first zeroth-order method to pretrain transformers to approach and sometimes even *exceed* backprop with large amounts of computation (bitter lesson!) - Dust is on the order of 1,000 to 10,000x more compute efficient than EGGROLL, the state-of-the-art ES method, for training transformers. - We chose pretraining transformers because it's arguably the hardest task possible for zeroth-order optimization. - Search in high dimensions is really misunderstood: zeroth-order usually gets better with overparameterization, not worse, i.e. at a fixed population size, larger models (up to 120x) reach a lower loss than smaller ones. - Backprop-like gradients emerge from a large population of activation perturbations, and the alignment holds up at every scale we tested, up to 1B tokens, which is critical for scaling. The core idea is a *virtual population*, which helps scale to large populations in transformers by perturbing activations in parallel. w/ @bishmdl76, @cs_serdar, @akshayvegesna