Photos and videos
Come to see the presentation of our work on Tokenizer Fairness at @emnlpmeeting. Today (Thu) at 17:45, session on "Multilinguality and Language Diversity 2" (A301).
1
28
Come to see the presentation of our work on Tokenizer Fairness at #EMNLP2025.
Today (Thu) at 17:45, session on "Multilinguality and Language Diversity 2" (A301).
1
1
1
121
Come to see the presentation of our work on Tokenizer Fairness at #EMNLP2025.
Today (Thu) at 17:45, session on "Multilinguality and Language Diversity 2" (A301).
1
1
1
121
15
supsi.ch/en/bando2022
Last few days to apply for a postdoc position in the NLP group @ IDSIA USI-SUPSI in beautiful Lugano.
nlp.idsia.ch/
122
IDSIA NLP retweeted
8 May 2024
Our preprint "Clustering of Disease Trajectories with Explainable Machine Learning: A Case Study on Postoperative Delirium Phenotypes" is out! A great work by @XiaochenZheng3 @manschuer Xingyu Chen and all co-authors 👏🎉!
Paper:
arxiv.org/abs/2405.03327
1
1
2
541
🤖 x 🧬=?
What happens when #generativeAI is trained on scientific texts from the literature, databases, genetic sequences, or code?
Explore its applications, and how our experts tackle its challenges in the #SIBProfile: issuu.com/sibswissinstituteo…
#EthicalAI #biocuration
2
7
9
954
Congratulations !
16 May 2024
Excited to announce our paper "StatBot Swiss: Bilingual Open Data Exploration in Natural Language" has been accepted to the ACL Findings 2024! 📝🎉 This joint work between @ZHAW, @swissstatistics, and @SDSCdatascience . Huge congrats to all involved! #ACL2024 @aclmeeting
58
IDSIA NLP retweeted
19 Jun 2024
Visit our poster at 11 am session
#NAACL2024
7 Mar 2024
Why more languages doesn't guarantee more linguistic diversity? How can we measure linguistic diversity in NLP benchmarks? Find out in our new pre-print:
arxiv.org/abs/2403.03909
@XimGutierrez @chrisbentz1 @stivits @olgapelloni
2
6
458
IDSIA NLP retweeted
21 Jun 2024
Our latest study "Machine learning prediction of prime editing efficiency across diverse chromatin contexts" is out @NatureBiotech! Shout out to our awesome collaborators @nicopmat @schwanklab @bvansteensellab and all (co)-authors 👏 nature.com/articles/s41587-0…
@UZH_Science @DQBM_uzh
21 Jun 2024
Our paper exploring the effect of chromatin on prime editing and diverse pegRNA designs was published today! 👨💻🧬🎉
rdcu.be/dLu0f
pridict.it
@UZH_Science @NatureBiotech @schwanklab @krauthammerlab @bvansteensellab
🧵
6
13
1,435
IDSIA NLP retweeted
20 Jun 2024
I was invited by @NatureNV to comment on a paper recently published in @Nature by @seb_far @janundnik @_lorenzkuhn @yaringal on "Detecting hallucinations in large language models using semantic entropy".
My piece on LLMs "Fighting fire with fire" is here:
rdcu.be/dLkVv
1
8
25
2,741
IDSIA NLP retweeted
24 Jun 2024
Very happy to have participated in the @AmericasNLP machine translation task this year
Great ambience, very dynamic community and beautifully careful with the diacritics in my name :)
2
3
25
1,906
A position available in our group:
supsi.ch/en/bando2022
A great opportunity to work on advanced NLP applications in a world leading research institute in Switzerland.
27
IDSIA NLP retweeted
8 Jan 2024
Excited to announce our paper "Exploring the Latest Highlights in Medical NLP across Multiple Languages: A Survey" published in the NLP section of the International Medical Informatics Association Yearbook 2023. Link here: thieme-connect.com/products/…
1
3
4
433
Great time to be in the NLP field.
15 Apr 2024
The 2024 AI Index tacitly shows Natural Language Processing rising to be the central technology of AI
2004: NLP way off in the AI margins
2014: A little excitement over chatbots
2024: AI Index leads with impressive progress of LLMs
aiindex.stanford.edu/report/
#NLProc #BiasedTakes
1
40
15 Jan 2024
Excited to be back in Japan for #BLAH8 where the focus this time is LLM usage in the context of biomedical annotation. Kicking things off we have @ProfLHunter making some great points about their evaluation
46
IDSIA NLP retweeted
15 Jan 2024
Excited to be back in Japan for #BLAH8 where the focus this time is LLM usage in the context of biomedical annotation. Kicking things off we have @ProfLHunter making some great points about their evaluation
1
11
892