Dr. Lucas Beyer’s slides

Full Transformer tutorial slides: http://lucasb.eyer.be/transformer

Main papers covered (at high level) in the computer vision talk:

Part1:

– VTAB: https://ai.googleblog.com/2019/11/the-visual-task-adaptation-benchmark.html

– Big Transfer: https://arxiv.org/abs/1912.11370

– Vision Transformer: https://arxiv.org/abs/2010.11929

– Are we done with ImageNet? https://arxiv.org/abs/2006.07159

– Scaling laws for shape optimization https://arxiv.org/abs/2305.13035

– Distillation of large models https://arxiv.org/abs/2106.05237

Part2:

– CLIP: https://openai.com/research/clip

– LiT (Locked image-text Tuning): https://arxiv.org/abs/2111.07991

– ImageBind: https://ai.facebook.com/blog/imagebind-six-modalities-binding-ai/

– PaLI image-text model: https://sites.research.google/pali/

– UViM: Unified Vision Model: https://arxiv.org/abs/2205.10337

– Unified loss (RL tuning): https://arxiv.org/abs/2205.10337