Guiseppe Carenini is giving a CODI keynote on discourse processing in the era of large language models #discourse #nlproc #acl2023nlp

When @aaronlolo326 first got in touch to discuss collaboration, I have to admit I was alarmed by some of his suggestions. I thought, *You can't just get rid of the world model!* But I was curious to see where his ideas might lead...

I'm very excited to announce our paper on "Functional Distributional Semantics at Scale", presented today at #StarSEM2023.

If you've ever wondered how truth-conditional semantics can really work when training on a large corpus, have a look!

#ACL2023NLP

An honour to be among those who are serving the ACL community! Thank you all for enabling all these *ACL events to happen (including those who were not present)! @aclmeeting #ACL2023NLP #NLProc

🥰 Martha Palmer, of lexical semantics in #nlproc fame and one of my profs at Penn, won the #ACL2023 lifetime achievement award. 💪
Seeing a female prof in computer science was hugely important for me 20 years ago (and continues to be now). #acl2023nlp

The keynote by @alisongopnik at #ACL2023NLP argues that we should think about LLMs differently: rather than (only) discussing about their intelligence, it is helpful to think of them as a new cultural technology that helps people accessing information.

Some personal notes on topics raised at the #ACL2023NLP panel on LLMs:
- NLP and Computational Linguists diverging
- Vertical applications becoming horizontal layers
- Role of open and closed source LLMs depending on usage
- Environmental concerns relevant
- Malicious actors using LLMs and the need for regulations
- NLP is becoming the natural science studying how LLMs work
- Academic research and research at tech companies enabling diversity (in academia) and scale (in industry)

#ACL2023NLP panel on LLMs: Dan Klein expresses enthusiasm for LLMs as representations of corpora for corpus linguistics. @meg adds only if the data is documented. 💯

At #acl2023nlp in Toronto, we present our work on directly retrieving facts from a knowledge graph without explicit entity & relation linking. Work with Jinheon Baek, Alham Fikri and Sung Ju Hwang.

Paper link: https://arxiv.org/pdf/2305.12416.pdf

#QuestionAnswering #LLMs #KnowledgeGraphs

Good morning Toronto! 🌆 Today at #ACL2023NLP you can expect the following presentations with participation of @UKPLab colleagues, starting at 09:00 in the Frontenac Ballroom and Queen's Quay.

https://arxiv.org/abs/2302.12611
https://arxiv.org/abs/2212.09885
https://arxiv.org/abs/2202.06690
https://arxiv.org/abs/2305.19597

Our team in Toronto (left to right): Haritz Puerto, Nils Dycke, Iryna Gurevych, Andreas Waldis and Yufang Hou represent the Ubiquitous Knowledge Processing (UKP) Lab at the 61st Annual Meeting of the Association for Computational Linguistics 🇨🇦 Follow them for more insights & impressions from the conference.

#NLProc #ACL2023NLP #MachineLearning

First day of #ACL2023NLP kicking off with Geoffrey Hinton giving interesting perspectives about the ability of #LLMs to understand and how goalposts have shifted in NLP.

Good morning Toronto 🇨🇦! Both events today at the 61st Annual Meeting of the Association for Computational Linguistics with participation of @UKPLab colleagues will take place at the Frontenac Ballroom and Queen's Quay and start at 11:00 and 14:00 respectively (CET 17:00 and 20:00). #ACL2023NLP

Update: A total of eight submissions with contributions by our colleagues have been accepted for the #ACL2023NLP! 🎉 Plus two submissions to this year's #SEM2023, which takes place co-located with the Annual Meeting of the ACL.

Hey #NLP—we all know better than to use NLP tools that we can't assess, right? These folks are advertising on the #acl2023nlp Rocket.chat, but give 0 info about how their system was built or evaluated.

The link for http://generalizable.xyz goes to a page that literally just consists of their logo.

In what sense is this GPT? What training data did they use? What test data? What evaluation metric?

But more to the pt: We're here to learn from each other, not to read synthetic versions of papers.

We're excited to have an opening for a postdoctoral researcher position in NLP with application in mental health coming up soon. The position is in beautiful Sydney! If you're attending @aclmeeting and wish to discuss, please reach out to me or otherwise feel free to email for more information! #NLProc #ACL2023NLP

#ACL2023NLP program is finally up: https://acl-org.github.io/ It's pretty full on! Looking forward to attending in person after four years! #NLProc @aclmeeting

Excited to share that our paper "Topic-Guided Sampling For Data-Efficient Multi-Domain Stance Detection" got accepted to #ACL2023

Work done with @isa @rnav_arora
#ACL2023nlp #NLP

While some of you are finalising your reviews, don't forget to nominate your favourite papers from 1998 or 2013 (10 years or 25 years ago!)! Call for nominations for ACL 2023 Test-of-Time (ToT) paper awards is only open till March 17: https://aclweb.org/portal/content/call-nominations-2023-test-time-tot-paper-awards #NLProc #ACL2023NLP

Hello there 👋! A gentle reminder that the #ACL2023NLP reviews are due on **March 10 (AoE; UTC - 12h)** ⏲️. If you haven't done so yet, please log into the system and check your assignments. Thank you in advance🙏.

Call for nominations for 2023 Test-of-Time (ToT) paper awards is inviting nominations for papers from 1998 or 2013. The deadline for nominations is March 17, 2023: https://www.aclweb.org/portal/content/call-nominations-2023-test-time-tot-paper-awards #NLProc #ACL2023NLP