AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
3min chapter
Robust Visual Reasoning with Adriana Kovashka - #463
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
How Do You Use Masking in a Multi Model Data Set?
In our earlier paper, which was on our ad state a set, it was not a transformer. We were using a agraph formulation to capture information from the image and from potential slogans in the image. So there we were using masking, not in the contects of transformers. And then for the v c r data set, we were using existing prior multi model transformer that had tokens from both the image in the text space and the question space. It's another we've talked about short cuts and masking. But often for popular brands, the method can cheat by using words that that aphear in the slogan.
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode