BREAKING
EgoBabyVLM Tests AI vs. Infants
Built by Meta FAIR and Partners
0hrs
BabyView recordings
0+
DevBench trials
How the Benchmark Works
1Egocentric video
2Object recognition
3Grammar & semantics
4Score grounding
Scaling Data vs. Learning Like a Baby
Big ModelsToday
Vast data and energy
Weak physical grounding
Infant LearningGoal
Limited first-person input
Embodied common sense
A Yardstick, Not a Solution
AI NEWS BLITZ
A new benchmark asks whether AI can learn from the world like a baby does.