#EMNLP2021livetweet #EMNLP2021 Generation by Splicing Together Nearest Neighbors - Sam Wiseman (Sam Wiseman), Arturs Backurs (Arturs Backurs), Karl Stratos (Karl Stratos) 👉 underline.io/events/192/ses…

I keep seeing published or under-review papers in #NLProc data-to-text generation using the RotoWire dataset where problems of train/test contamination are not addressed. Evaluation are also poor. I created a simple post to explain both including a TL;DR github.com/nlgcat/sport_s…

🚨📑🤖🚨 Excited to announce that the Automatic Summarization for Creative Writing Workshop (#CreativeSumm) will take place at #COLING2022! For updates on the timeline check: creativesumm.github.io

🚨🚨Registration for the COLING 2022 #CreativeSum Workshop Shared Tasks (books & tv/movie script summarization) is now open, the form is available here: bit.ly/3LFBXC4 Registration closes July 1st. #NLProc

And this cool paper from 2021 which deserves more recognition arxiv.org/abs/2102.13249

As Apple Intelligence is rolling out to our beta users today, we are proud to present a technical report on our Foundation Language Models that power these features on devices and cloud: machinelearning.apple.com/research/apple…. 🧵