At 12:15am today, the #NeurIPS2023 Google booth will host Yonatan Bitton for a demo of SeeTRUE, a tool for testing the alignment between text and images that can pinpoint areas of mismatch and can help adjust ranking in text-to-image generation tasks.
AI
-
SAS Innovate 2026: Call for AI and Analytics Speaker Submissions
By
–
Becoming a speaker at SAS Innovate is just the icing on the cake . Don’t miss the opportunity to be surrounded by AI and analytics experts, just like you. Get those content suggestions in by Dec. 15 . http://
2.sas.com/6017RHhXJ #ArtificialIntelligence #SASInnovate -
BPE Tokenization Dies: 2024 Marks Major Transformer Evolution
By
–
i can see it now: 2024 will be remembered as the year BPE died tokenization is by far the clunkiest part of a transformer; one last remaining bit of inelegance in an otherwise hyperoptimized model architecture time for it to go
-
ChatGPT’s First Year: AI Evolution Since Launch
By
–
It's hard to believe, but ChatGPT is now over a year old. I tried to put a blog post about what has happened in the World of AI since then, and I quickly realized that a comprehensive overview would be beyond anyone's purview. Nonetheless, I managed to put together a piece on
-
DALL·E and the missing text in the image
By
–
The text was added in post, DALL E couldn’t drop the caption into the frame – it wrote it in a separate prompt referencing the image.
-
Stanford AI Tool Enables Composers to Create Music with AI Control
By
–
Developed by scholars from @StanfordCRFM @StanfordHAI
, a new tool called the Anticipatory Music Transformer enables composers and musicians to craft melodies alongside AI with more control over the creative process. https://
stanford.io/4atgadV -
Difficulties in generating humorous captions
By
–
Not that I’m aware of. It struggled to drop the caption into the image, so that part I had to add in post. It took about 40 iterations to get a caption that I thought was funny, so one-shot results would be hit or miss. Someone should make one though!
-
DreaMoving: AI Video Generation with Face Transfer and Pose Control
By
–
This looks interesting – like AnimateAnyone and MagicAnimate, but with prompting and face transfer.
— fofr (@fofrAI) 11 décembre 2023
DreaMoving:
reference image + pose sequence + prompthttps://t.co/Ey45v2g7JY pic.twitter.com/sqpW3sSm3hThis looks interesting – like AnimateAnyone and MagicAnimate, but with prompting and face transfer. DreaMoving:
reference image + pose sequence + prompt https://
dreamoving.github.io/dreamoving/ -
Google Platinum Sponsor at NeurIPS2023 Workshop Discusses Career Opportunities
By
–
Google is also excited to be a Platinum Sponsor of today's #NeurIPS2023 @_LXAI workshop. Attending the workshop? Hear panelist Theo Guidroz at 1:15pm or talk to any of several other Googlers in attendance to learn all about life at Google! [2/2]
-
Google Sponsors NeurIPS 2023 Women in ML Workshop
By
–
Google is proud to be a Platinum Sponsor of the #NeurIPS2023 @WiMLworkshop happening today. Curious about life at Google? Visit with Lara McConnaughey during the mentorship hour, or speak with one of the many other Googlers in attendance! [1/2]