Since the launch of #GPT4 in March, researchers worldwide have been exploring its capabilities. So have our colleagues Tilman Beck (Tilman Beck) and Martin Tutek (Martin Tutek) at UKP Lab, generously supported by a Microsoft Azure for Research Grant which has just been extended! 🎉🎉🎉

Thanks a lot for the invitation🙏, I really enjoyed this retreat. It was a great mix of insightful presentations, discussions and activity.

Are Emergent Abilities in Large Language Models just In-Context Learning? Spoiler: YES 🤯 Through a series of over 1,000 experiments, we provide compelling evidence: arxiv.org/abs/2309.01809 Our results allay safety concerns regarding latent hazardous abilities. A🧵👇 #NLProc

People ask me all the time if I am "pro-Israel" because I am a Jew who has lived in Israel, and my answer is that being "pro-Israel" or being "pro-Palestine" or being a "Zionist" does not properly capture the nuance of thought most people do or should have about this issue. It

Just logged into my OpenReview account and noticed that the papers from the ARR October cycle have been assigned with a first (small) task due Saturday. However there was no notification at all 🫤

Just realized that deactivating chat history (e.g. bc you don't want OpenAI to store & use your data for model training) also deactivates all of the fancy GPT-4 functionalities, such as plugins or code interpreter. I guess the $20 subscription is only a part of your "payment" :/

Just #ARR things #NLProc

Followers from obvious fake-accounts has massively increased for me over the last weeks. Anybody else experiencing this?

Seven papers authored or co-authored by UKP staff have been accepted for publication at this year's eaclmeeting! Congratulations to all authors – see you in Malta 🇲🇹! #EACL2024 #NLProc informatik.tu-darmstadt.de/ukp/ukp_home/u…

How long is the maintenance going to last? 🧐 Hugging Face

daily dose of tokenizer weirdness #NLProc

Ever faced a lack of labeled data for multilingual tasks? This #EACL2024, we unveil an effective method for sentiment analysis in low-resource languages solely relying on a multilingual lexicon 💡 – more in this 🧵 (1/8). 📰 arxiv.org/abs/2402.02113 #NLProc #SentimentAnalysis

LLMs are increasingly prompted with different user profiles to solve subjective NLP tasks. What are the factors which determine what the model generates? Discover it in our #EACL2024 paper – learn more in this 🧵 (1/8). 📰 arxiv.org/abs/2309.07034 #NLProc #Prompting

We are proud to announce that the contribution »Sensitivity, Performance, Robustness: Deconstructing the Effect of Sociodemographic Prompting« by Tilman Beck, Hendrik Schuff, Anne Lauscher (she/her) (Universität Hamburg) and Iryna Gurevych (UKP Lab) has just been awarded the #EACL2024 Social Impact Award!

Congrats to all my co-authors and especially, to Tilman Beck for receiving this recognition! Thank you, eaclmeeting !

Wow, thanks a lot for the appreciation of our work on sociodemographic prompting at #eacl2024!

Good at language 🟰 good at thinking? Not true for LMs! Even if we might think that when talking to them 💬 Meet Holmes 🔎, a benchmark to assess linguistic competence📚 of LMs untwined from other skills 🔥 Spoiler 🤫 Architecture is one key! (1/🧵) 🌐 holmes-benchmark.github.io

Stop complaining about the bad review quality. Join forces and start research on #NLProc for #PeerReview! 🚨 A new white paper by over 20 top AI and NLP researchers provides a thorough discussion of AI assistance for scientific quality control. (1/🧵) 📑 arxiv.org/abs/2405.06563

I shared a controversial take the other day at an event and I decided to write it down in a longer format: I’m afraid AI won't give us a "compressed 21st century". The "compressed 21st century" comes from Dario's "Machine of Loving Grace" and if you haven’t read it, you probably