E2E encryption and decentralization are natural tools to create software robust against malicious service providers. Our work with @inkandswitch presents BeeKEM, an efficient decentralized group key agreement protocol provable security, and a performant implementation.
@itsandrewgao As perhaps expected a model trained on language A may do better or worse on different language B's, but these particular relationships are hard to explain: they don't seem typological or etymological in nature
Can you reliably get image generation models like DALL-E 2 to illustrate specific visual concepts using made-up words? In this new preprint, I show that you can, using new approaches for text-based adversarial attacks on image generation. 1/12
https://t.co/fR8el9TLAr