记录|Celeste

2022

2022-01-26
ALIGN: Contrastive Vision + Language Representation Learning with Noisy Text Supervision
2022-01-26
Something-Else: Compositional Action Recognition
2022-01-18
Ego4D Dataset: Advancing Multimodal Perception of Egocentric Video
2022-01-16
Neural Module Network: Compositional ViQA Network Built on the Fly
2022-01-12
Language as the Unified Protocol for Generalization
2022-01-10
MetaFormer: Token Mixer is What You Need for Transformer
2022-01-07
Dall-E: Zero-Shot Text-to-Image Generation
2022-01-06
SCAN, gSCAN, ReaSCAN: Benchmark Model's Compositional Generalization Skills
2022-01-06
Vision Transformer: Image is Worth 16x16 Words
2022-01-05
CLIP: Connecting Text & Images