Andrew White(@andrewwhite01) 's Twitter Profileg
Andrew White

@andrewwhite01

Head of Sci/cofounder @FutureHouseSF. Prof of chem eng @UofR (on sabbatical). Wrote textbook on deep learning in chemistry (https://t.co/WXOkMs9pkB).

ID:31032895

linkhttp://thewhitelab.org calendar_today14-04-2009 02:25:55

5,9K Tweets

20,3K Followers

1,5K Following

TheRealAmoo(@MehradAnsari) 's Twitter Profile Photo

In our latest work, Andrew White and I show how you can leverage the positive class only in imbalanced peptide datasets and train a binary classifier for different antimicrobial activities.

Full paper: pubs.rsc.org/en/content/art…

account_circle
Andrew White(@andrewwhite01) 's Twitter Profile Photo

We run unit tests on various embeddings types as part of our CI. One of the unit tests is checking consistency of the embedding model. OpenAI embeddings comeback with a few percent difference each time. Is that expected? Seems weird

account_circle
Andrew White(@andrewwhite01) 's Twitter Profile Photo

Anyone know where I can buy a framed CI check mark that looks like the github one? I need a physical green check mark

account_circle
FutureHouse(@FutureHouseSF) 's Twitter Profile Photo

Join us at FutureHouse on Zoom tomorrow, Apr 16th at 10:00am for a FutureTalk featuring guest speaker Adrienne Hoarfrost. Fill out the form below to gain access to her talk on using AI to Illuminate 'Microbial Dark Matter'
airtable.com/appYoFRg45Cub7…

Join us at FutureHouse on Zoom tomorrow, Apr 16th at 10:00am for a FutureTalk featuring guest speaker Adrienne Hoarfrost. Fill out the form below to gain access to her talk on using AI to Illuminate 'Microbial Dark Matter' airtable.com/appYoFRg45Cub7…
account_circle
Jablonka Lab (Lab for AI for Materials)(@jablonkagroup) 's Twitter Profile Photo

Are large language models superhuman chemists?

We have been wondering about this, too and developed the first automated and comprehensive benchmark for evaluating large language models in chemistry and let LLMs compete with human chemists 👩🏼‍🔬

📝 arxiv.org/abs/2404.01475

Are large language models superhuman chemists? We have been wondering about this, too and developed the first automated and comprehensive benchmark for evaluating large language models in chemistry and let LLMs compete with human chemists 👩🏼‍🔬 📝 arxiv.org/abs/2404.01475
account_circle
Andrew White(@andrewwhite01) 's Twitter Profile Photo

Does anyone have a snippet for downloading a PDF from a chemrxiv DOI link? They use a lot of javascript to get the download button.

account_circle
Andrew White(@andrewwhite01) 's Twitter Profile Photo

Why do so few institutions have access to Nature Machine Intelligence? Is it on a separate contract than usual for libraries. Seen it missing at a few major university libraries.

account_circle
Andrew White(@andrewwhite01) 's Twitter Profile Photo

Where do people get the raw data for estimating biotech pipelines? Is it from websites? Would love to see raw data for companies, modalities, targets pre-clinical.

account_circle
Eric Schmidt(@ericschmidt) 's Twitter Profile Photo

I visited FutureHouse this week, and was amazed by the progress Sam Rodriques, Andrew White, and the others are making. They are extremely fast. A new way of doing science!!!

account_circle
Andrew White(@andrewwhite01) 's Twitter Profile Photo

I really want to share demos on social media of stuff we make at FutureHouse , but the compression just obliterates anything with text. We should just pivot to being a robot only company so our demos look cooler.

account_circle
Andrew White(@andrewwhite01) 's Twitter Profile Photo

Mildly concerned that you can take a URL of a person's face and use them in a midjourney prompt. Here's me smoking, drinking, and riding a dinosaur from one photo

Mildly concerned that you can take a URL of a person's face and use them in a midjourney prompt. Here's me smoking, drinking, and riding a dinosaur from one photo
account_circle