Stories by anima

We keep making transformers faster. What if we don't need them at all?

A One-Shot Catastrophe Avoidance Benchmark for RL Agents

Semantic Field Execution: Decoupling Transformers from Runtime Inference

The Inference Efficiency Paradox

Post-transformer inference: 224× compression of Llama-70B with improved accuracy

MIT PrivateKit and Safe Paths Webinar COVIDathon

Universal Basic Robots aka Decentralized Autonomous Workers

Why a robot tax won't work [list]

CooperVision debuts contact lenses designed to ease screen fatigue

Superintelligence, our Superdoom [10 min read]

Analysts are overlooking how much AR and VR are about to change our lives

Glass Ceiling for European VR+Drone Startups --|

How Vibrant is the Hardware Startup World