📝 New from FAIR: An Introduction to Vision-Language Modeling. Vision-language models (VLMs) are an area of research that holds a lot of potential to change our interactions with technology, however there are many challenges in building these types of models. Together with a set

Will you be at #CVPR2024 ? If so, join us for the Responsible GenAI Workshop on Tuesday morning at 8:30am! We have a great set of speakers and panelists, as well as a poster session at 10:00am! Check out the full program for more: sites.google.com/view/cvpr-resp… #CVPR2024

We're excited to announce our detailed schedule featuring amazing speakers (Besmira Nushi 💙💛 Miranda Bogen @Björn Ommer) and panelists (Alex Beutel Milagros Miceli Cristian Canton Chengzhi Mao David Bau ). Huge thanks to Google for their sponsorship! See you #CVPR2024 in Summit Room 433 on 18th!

🚨Excited to share our recent work on improving the diversity of generations in diffusion models. TL;DR: In this work we apply additional guidance by computing the Vendi Score of previously generated images and enforce the new generations to be different from previous ones (1/4)

If you're at #CVPR, come by our poster tomorrow at the ReGenAI Workshop from 10:00 to 10:45.

Are sota image generative models effective world models? Consistency-diversity-realism Pareto fronts show they're not (yet): - No model dominates others as a world model - Improvements in quality and consistency have come at the expense of diversity 🔗 arxiv.org/abs/2406.10429

Had a wonderful time working with this great team of students and fellow mentors!!

Can generative image models be good world models? This work from Meta FAIR shows that there is a tradeoff between realism and diversity. The more realistic a generative model becomes, the less diverse it becomes. Realism comes at the cost of coverage. In other words, the most

Super excited for Contextualized Vendi Score Guidance being accepted at #ECCV2024 🎉 The codebase is also available now: github.com/facebookresear… c-VSG can be applied on any off-the-shelf latent diffusion model to improve the diversity of generations. arxiv.org/abs/2406.04551

We will be presenting our work on improving the diversity of text-to-image generative models at #ECCV2024 on Tuesday at 10:30a, Poster #108!