Maybe at the Nobel lecture we'll finally find out how many legs Hinton's cat really has... #acl2023 #throwback @7c0h https://www.nobelprize.org/prizes/physics/2024/press-release/

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

Fantastic keynote by Alison Gopnik at #acl2023nlp , arguing that we should see large language models as a new cultural technology (think printing press, internet search) and not new kinds of (intelligent or not) agents #ACL2023 #llm

See the sheer joy of my collaborators at #ACL2023 when
DissentQA
won best Paper AC award

This is a happy outcome of the fruitful collaboration with a group of wonderfully friendly people

https://arxiv.org/abs/2211.05655
#nlproc #machinelearning #Qa #factuality

Presenting Riveter , a Python package to measure social dynamics between personas mentioned in text.

Given a verb lexicon, Riveter can extract entities and visualize relationships between them.

Package: https://github.com/maartensap/riveter-nlp

Paper: http://maartensap.com/pdfs/antoniak2023riveter.pdf

Video: https://youtube.com/watch?v=Uftyd8eCmFw

Demo Notebook: https://github.com/maartensap/riveter-nlp/blob/main/riveter/demo.ipynb

With Anjalie Field, Jimin Mun, Melanie Walsh, Lauren Klein, Maarten Sap

Check out our new paper "DARE: Towards Robust Text Explanations in Biomedical and Healthcare Applications", oral at @aclmeeting by lead author Adam Ivankay this Wednesday!

We show adversarial attacks to #explainability methods for #DeepNeuralNetworks in technical text domains, propose a quantification of this problem, and initial solutions.

Presentation: https://virtual2023.aclweb.org/paper_P1265.html
Paper: https://arxiv.org/abs/2307.02094
Code: https://github.com/ibm/domain-adaptive-attribution-robustness

So cool to see that #acl2023 is still the place where I can meet many old friends #computationalLinguistics

Die Bochumerinnen fühlen sich jedenfalls gleich wie zuhause #Toronto #Acl2023 #industrieruine

This is great news from ACL 2023:

https://2023.aclweb.org/registration/discounted_virtual_registration/

Discounted virtual registration ($0-100) for people attending from regions where ACL registration fees are out of reach.

Please spread the word! Applications due June 28.

Excited to share our new #acl2023 Findings paper! #nlp #nlproc

We introduce NeQA, a dataset consisting of questions with negation, in which language models exhibit inverse, U-shaped, or positive scaling, depending on prompting methods or model families.

https://arxiv.org/abs/2305.17311

Do I have any friends in #Toronto #TO who I could stay with for a couple days of the #ACL2023 workshops? I arrive the 12th and leave the 14th so just two nights of my dumb ass. I'll buy you any snacks you desire

How can we explore hidden biases in language models impacting fairness? Our #ACL2023 demo paper introduces Finspector, an interactive visualization widget available as a Python package for Jupyter, that helps uncover these biases.

Paper, Video, Code: https://www.bckwon.com/publication/finspector/
arXiv: https://arxiv.org/abs/2305.16937
GitHub: https://github.com/IBM/finspector

Our paper showcases a use case and discusses implications, limitations, and future work.

What do we learn from modeling second language acquisition (SLA)? Read our paper for #ACL2023 to find out about the importance of *negative* transfer + which elements of child-directed speech do and don't survive in text-based language models + a bonus new multi-lingual CDS corpus.

https://arxiv.org/abs/2305.19589

Two UKP papers have been accepted to *SEM 2023, which will happen July 13-14 as part of #ACL2023! Read the pre-prints here:
https://arxiv.org/abs/2211.01874
https://arxiv.org/pdf/2205.06733

Congratulations to the authors Tilman Beck, Andreas Waldis, Dominic Petrak, Nafise Sadat Moosavi and Iryna Gurevych!

#SEM2023 @sheffielduni @hslu_luzern

Yay! 2/2 papers accepted at #ACL2023 !

First, in the main conference, there's
"FIREBALL: A Dataset of Dungeons and Dragons Actual-Play with Structured Game State Information" by @zhuexe, Karmanya Aggarwal, Alex Feng, myself, & @ccb

Contributions:
- A corpus of data from people playing #DnD on Discord using a bot called #Avrae, made by @zhuexe himself. Avrae tracks vital game state information for D&D.
- #LLMs "translating" Avrae commands into plain English.

arXiv link TBA!
1/2

We need to re-think in-person conferences. Is hybrid alone enough? What about local satellites? #acl2023 #assets2023 #participation Visa processing times exclude people from conferences

https://2023.aclweb.org/blog/visa-info/

Writing the "Limitations" section for an #ACL2023 submission feels a bit like scripting the text for Reviewer 2.

#acl2023 early bird: Have we in the #NLP #NLProc community resolved the century-old author order puzzle? Who should be the last author? The most senior author or the author least contributing actual experimenting and writing?