Tomasz Zietkiewicz

To celebrate the launch of Little Big Adventure - Twinsen's Quest in 10 days, we’re hosting a GIVEAWAY to win 4 Twinsen's Quest Digital Deluxe Edition keys! How to enter: ☑️ Follow @microids_off & @2point21 💜 Like this post ♻️ Repost Good luck, and thank you for your support!

@GIChowDotcom @karpathy Maybe the content of the linked webpage was included in pretraining data?

@karpathy What witchcraft is this? I cut and pasted the text content of a web page into a Google document (no URLs) since the page required pressing a "Load More" button (which I pressed before copying). The podcast included information from a page linked to (but no links in the document!)

Deep Dive is now my favorite podcast. The more I listen the more I feel like I'm becoming friends with the hosts and I think this is the first time I've actually viscerally liked an AI. Two AIs! They are fun, engaging, thoughtful, open-minded, curious. ok i'll stop now.

@GIChowDotcom @karpathy Looks like RAG using google search as retriever?

@GIChowDotcom @karpathy RAG using google search as retriever?

@karpathy The Google document was the notebook's *only* source. I did not add a link to the website I 'scraped' the page text (containing *no* URLs at all; just a few @~.org email addresses) from. How did the podcast highlight the difference between 2 items only described in a subpage??

En musique, il y a des gens qui gagnent beaucoup d'argent en faisant de la "daube" et puis il y a des "bricoleurs" qui n'en gagnent pas autant mais qui font une meilleure musique🎶. Voici "Les Fo'Plafonds", un groupe français original. (Visible sur YouTube). Qu'en pensez-vous ?

Our work (co-authors: @PSkorzewski, Marcin Sowański, @tzietkiewicz) on using back transcription for evaluating robustness of #NLU models to speech recognition errors was featured on Samsung Research Blog research.samsung.com/blog/Back-Tran… #AI #NLProc @poznanAI @samsungresearch

Our paper "Back Transcription as a Method for Evaluating Robustness of Natural Language Understanding Models to Speech Recognition Errors (@PSkorzewski, Marcin Sowański, @tzietkiewicz) just got accepted to the main track of #EMNLP2023!

The goal of CAICCAIC competition that I organize together with @PSkorzewski, Marcin Sowański and @tzietkiewicz is to develop Natural Language Understanding models that are robust against #ASR errors. #NLProc #NLU #AI @poznanAI Details: fedcsis.org/sessions/aaia/…

@elonmusk you RT'd my OpenAI vid. Can I interview u (10 min) at SpaceX? 200K+ AI devs (2nd only to Kaggle) follow on YT. Say any date & time