Hello everyone! I will be presenting today at #acl2022nlp (Poster Session 6: Machine Translation and Multilinguality starting from 10:45 AM local time) our work on 'Multi-Task Learning for Zero-Shot Performance Prediction' (w/ Shanu Kumar , Sandipan Dandapat, Monojit Choudhury )↓ (1/n)
☘️ Happy to attend ACL 2023 and present our long paper: 'Slangvolution: A Causal Analysis of Semantic Change and Frequency Dynamics in Slang' with @Daphiya Zhijing Jin and Mrinmaya Sachan from ETH Zurich and Intelligent Systems #acl2022nlp #NLProc
Do you want to know how to best explain your #HateSpeech detector? Come see our poster NLP Power! workshop at #acl2022nlp ☘️
When: Today, Thu 26 at 11
Where: Gather Town #17
Debora Nozza Eliana Pastor Dirk Hovy
#ACL2022
.Dongyeop Kang (DK) is moderating a panel discussion at #In2Writing on bridging NLP & HCI to design, build, & evaluate writing assistants, with Melissa Roemmele (@melissa_roem), Courtney Napoles (@courtnap), Qian Yang (@fabulousQian), & Sherry Wu (@tongshuangwu). #ACL2022 #acl2022nlp
#ACL2022NLP students and recent graduates: if you’ve never been to a conference, have fun! Introduce yourselves, don’t be shy :)
Check out our paper
“Retrieve Fast, Rerank Smart: Cooperative and Joint Approaches for Improved Cross-Modal Retrieval“
An efficient retrieve and rerank approach for image search!
w/ Gregor Geigle Nils Reimers Ivan Vulić
Paper: arxiv.org/pdf/2103.11920…
Code: github.com/UKPLab/MMT-Ret…
Can pretrained language models memorize and exploit contaminated downstream test sets?
📣Excited to share our paper 'Data Contamination: From Memorization to Exploitation' with Roy Schwartz.
arxiv.org/abs/2203.08242
#acl2022nlp
TL;DR 🧵👇
(1/8)
So excited! on the ✈️ to Dublin for my first ever conference, #acl2022nlp ! two papers I co-authored will be presented and I can’t wait to tell you about them! 🧵1/n
Meet LAGr, a sequence-to-graph semantic parser that generalizes more systematically than prior seq2seq approaches.
If you're at #acl2022nlp , come see our poster being presented right now at #47 !
Paper: arxiv.org/pdf/2205.09607…
Had a great time (and some very felt discussions 🙃) at the meet-up for #HateSpeech researchers ACL 2023!
Thank you everyone for showing up!
#acl2022nlp
You know you're at a CS conference in Europe when you overhear this in the stairwell: 'Huh, why is this hotel zero indexed?' #acl2022nlp
📢 Last-minute #acl2022nlp paper ad 📢 Have you ever wondered whether there might be a simple way to fix some of the weird tokenizations produced by the likes of BPE and WordPiece? Don't look any further - our new paper has you covered. A 🧵 1/
Our 'AraT5: Text-to-Text Transformers for Arabic Language Generation' at ACL 2022 main conference aclanthology.org/2022.acl-long.…
#acl2022nlp