Meta FAIR recently released the Seamless Interaction Dataset, the largest known high-quality video dataset of its kind, with: 4,000+ diverse participants
4,000+ hours of footage
65k+ interactions
5,000+ annotated samples This dataset of full-body, in-person, face-to-face
@aiatmeta
-
Meta FAIR Releases Seamless Interaction Dataset for Video Research
By
–
-
Meta releases open source AI tool for sustainable concrete discovery
By
–
This week we shared an open source AI tool that will help accelerate the discovery of high-performance, low-carbon concrete. See the full story in the thread below. You can also find research artifacts for this project here: Technical report with details of the model and
-
Mark Discusses Meta’s Vision for Personal Superintelligence
By
–
"Our mission with the lab is to deliver personal superintelligence to everyone in the world. So that way, we can put that power in every individual's hand." – Mark Watch Mark's full interview with The Information as he goes deeper on Meta's vision for superintelligence and
-
Mark Announces Meta’s Major AI Compute Investment
By
–
Today Mark announced Meta's major AI compute investment. See his post: https://
facebook.com/share/v/1AnKhQ
ouDb/
… -
Seamless Interaction and Natural Conversational Dynamics
By
–
Head to our blog to go deeper on the full story: https://t.co/KUx2yBSpAU pic.twitter.com/txKkW04hCd
— AI at Meta (@AIatMeta) 27 juin 2025Head to our blog to go deeper on the full story: https://
ai.meta.com/blog/seamless-
interaction-natural-conversational-dynamics/?utm_source=twitter&utm_medium=organic_social&utm_content=video&utm_campaign=seamlessinteraction
… -
Meta Releases Seamless Interaction Dataset with 4000+ Hours Video
By
–
We’re also publicly releasing the Seamless Interaction Dataset, with 4,000+ participants and 4,000+ hours of interactions, making it the largest known video dataset of its kind.
— AI at Meta (@AIatMeta) 27 juin 2025
This dataset enables our audiovisual behavioral models to understand and generate human-like social… pic.twitter.com/TZjLC5jZ7rWe’re also publicly releasing the Seamless Interaction Dataset, with 4,000+ participants and 4,000+ hours of interactions, making it the largest known video dataset of its kind. This dataset enables our audiovisual behavioral models to understand and generate human-like social
-
Meta FAIR Introduces Seamless Interaction for Interpersonal Dynamics Modeling
By
–
🚀New from Meta FAIR: today we’re introducing Seamless Interaction, a research project dedicated to modeling interpersonal dynamics.
— AI at Meta (@AIatMeta) 27 juin 2025
The project features a family of audiovisual behavioral models, developed in collaboration with Meta’s Codec Avatars lab + Core AI lab, that… pic.twitter.com/vptAnJwkoVNew from Meta FAIR: today we’re introducing Seamless Interaction, a research project dedicated to modeling interpersonal dynamics. The project features a family of audiovisual behavioral models, developed in collaboration with Meta’s Codec Avatars lab + Core AI lab, that
-
Meta releases technical report on dyadic audiovisual motion modeling
By
–
One more thing… you can also check out this technical report detailing our methodology to build motion models on the dataset, along with an evaluation framework for this type of model: https://
ai.meta.com/research/publi
cations/seamless-interaction-dyadic-audiovisual-motion-modeling-and-large-scale-dataset/?utm_source=twitter&utm_medium=organic_social&utm_content=video&utm_campaign=seamlessinteraction
… -
Emmanuel Dupoux Receives Prestigious Award for Language Modeling Research
By
–
Huge congratulations to Emmanuel Dupoux on receiving this prestigious award! We're honored to have collaborated with Emmanuel on groundbreaking research in language modeling & developmental AI. Thank you to @CNRS_Paris for recognizing his outstanding contributions.
-
Meta Announces First Llama Startup Program Cohort Selection
By
–
The response to our first-ever Llama Startup Program was astounding, and after reviewing over 1,000 applications we’re thrilled to announce our first group. This eclectic group of early-stage startups are ready to push the boundaries of what’s possible with Llama and drive