📢New research🧵How can we estimate the quality of our models for AGI tasks?
1/ My latest paper dives into this fundamental epistemic question. The core finding? Under current data collection practices, we *cannot* know a model's quality for key AI tasks 😬
Today we're releasing Community Alignment - the largest open-source dataset of human preferences for LLMs, containing ~200k comparisons from >3000 annotators in 5 countries / languages!
There was a lot of research that went into this... 🧵
📑 Get the full paper here: https://t.co/Acjg1SPNk6
👋 Make sure to say Hi at NeurIPS if you are around and want to discuss more
🙏 Many thanks to Leon Bottou, Tina Eliassi-Rad, @SmithaMilli, @kchonyc for valuable feedback!
📢New research🧵How can we estimate the quality of our models for AGI tasks?
1/ My latest paper dives into this fundamental epistemic question. The core finding? Under current data collection practices, we *cannot* know a model's quality for key AI tasks 😬
These results are very related to @ylecun's recent point on the challenge of long tails in advancing AI research. However, the paper looks at it from a validation perspective, not a learning perspective.
🎥 Today we’re premiering Meta Movie Gen: the most advanced media foundation models to-date.
Developed by AI research teams at Meta, Movie Gen delivers state-of-the-art results across a range of capabilities. We’re excited for the potential of this line of research to usher in entirely new possibilities for casual creators and creative professionals alike.
More details and examples of what Movie Gen can do ➡️ https://t.co/M19x2ndwnr
🛠️ Movie Gen models and capabilities
Movie Gen Video: 30B parameter transformer model that can generate high-quality and high-definition images and videos from a single text prompt.
Movie Gen Audio: A 13B parameter transformer model that can take a video input along with optional text prompts for controllability to generate high-fidelity audio synced to the video. It can generate ambient sound, instrumental background music and foley sound — delivering state-of-the-art results in audio quality, video-to-audio alignment and text-to-audio alignment.
Precise video editing: Using a generated or existing video and accompanying text instructions as an input it can perform localized edits such as adding, removing or replacing elements — or global changes like background or style changes.
Personalized videos: Using an image of a person and a text prompt, the model can generate a video with state-of-the-art results on character preservation and natural movement in video.
We’re continuing to work closely with creative professionals from across the field to integrate their feedback as we work towards a potential release. We look forward to sharing more on this work and the creative possibilities it will enable in the future.
Make sure to stop by the excellent contributed talks as well as the poster session too!
Co-organized with Alessandro Lazaric, @ArpitAgarwalAI, @HodaHeidari, Nicolas Usunier, and @tinaeliassi
Workshop website: https://t.co/WbtBaQRPmn
I had a wonderful time attending and presenting @FAccTConference last week. To reflect on the experience and sum up what I learned, I wrote the following blog post, featuring data work, Algorithmic Impact Assessments, and hotpot (thanks to @angelamczhou!)
https://t.co/lhqAVokx79
I'm very happy that I had the chance to collaborate with @dayvidliu@mnick (and twitterless Nico Usunier, as always!) on a #FAccT2023 paper on group fairness without demographics, a topic I particularly care about 😌
https://t.co/NPQSuvYHKp
@MetaAI
📣 A new #ICML2023 paper investigates the Kinetic Energy of Gaussian Probability Paths which are key in training diffusion/flow models. A surprising takeaway: In high dimensions *linear* paths (Cond-OT) are Kinetic Optimal!
Led by @shaulneta w/ @RickyTQChen@lematt1991@mnick
Excited to share our new work on Riemannian Flow Matching.
Unlike diffusion-based approaches, it’s
- completely simulation-free on simple manifolds,
- trivially applies to higher dimensions,
- tractably generalizes to general geometries!
https://t.co/sO2nSQrTjZ
w/ @lipmanya
@lipmanya@RickyTQChen@helibenhamu@mnick@lematt1991 I wanted to check how Flow Matching-OT worked in practice and ... it is GREAT! It is very easy to implement and trains super fast. Here is a PyTorch demo in only a 100 lines of code: https://t.co/ViA3EMIftt
**Flow Matching** (#ICLR2023 spotlight) offers a simple simulation-free method for training flow-based generative models, generalizing and improving upon diffusion models in training speed, sampling efficiency, and generation quality. @RickyTQChen@helibenhamu@mnick@lematt1991