Ideas and concepts that are the result of overly precise science and intense thought frequency are those that, to many people, at first glance look like it is a "religion".
Conversely, those that don't look like that are generally no big deal.
Summary of great research progress on #GoogleAI, including language models, computer vision, multimodal models, generative ML. We're building it all into current and upcoming products + APIs, look forward to sharing more with everyone soon. Stay tuned!
https://t.co/bGzLSfxZ0A
Happy 7th birthday @OpenAI! Most people know about this company through DALLE & ChatGPT.
I interned at OpenAI in 2016-17 on the “Universe” initiative, which inspired my Minecraft work @NVIDIAI. In its early days, OpenAI hit 7 cool milestones in games, agents & robotics.
Long🧵:
Parsel: A Unified Natural Language Framework for Algorithmic Reasoning
https://t.co/QMsGYdYUq8
Parsel is a framework enabling automatic implementation and validation of complex algorithms with code LLMs. Parsel can...
🧵 👇
Why does ChatGPT work so well? Is it “just scaling up GPT-3” under the hood? In this 🧵, let’s discuss the “Instruct” paradigm, its deep technical insights, and a big implication: “prompt engineering” as we know it may likely disappear soon:👇
Why building the metaverse is hard:
Games startups are more like a space tech or biotech startups than building an app
To build a v1 of a new game usually takes 2-3 years and tens of millions of dollars at the low end
Apps take 6 months and <$1m
StructDiffusion constructs structures out of a single RGB-D image based on high-level language goals. The model trained in simulation generalizes to real-world objects the robot has never seen before.
💪with Tucker, Sonia, and @chris_j_paxton
🔗: https://t.co/ekXV29yu50
👇🧵