Oslo, Norway
Joined October 2009
- Tweets113
- Following232
- Followers207
- Likes128
Stephan Oepen retweeted
Grateful that my #acl2020nlp theme paper "The Unstoppable Rise of Computational Linguistics in Deep Learning" was accepted. Traces how the nature of language has and will impact neural network architectures, including variable binding in Transformer. arxiv.org/abs/2005.06420
Stephan Oepen retweeted
In case you haven’t heard, the new unit for measuring computation runtime is TPU core years. But, if you missed that memo, since the numbers are already in the hundreds, you may as well get ahead of the game and start quoting your runtimes in TPU core centuries #NLProc
unavailable
Cross-Framework and Cross-Lingual Meaning Represenation Parsing: AMR, DRG, EDS, PTG, and UCCA—all in a uniform package and evaluation framework. Come play at the CoNLL 2020 shared task, and submit parser outputs by August 10! mrp.nlpl.eu
Stephan Oepen retweeted
Dear all, upon the 1st call, a vivid and important discourse has emerged. Please find our statement here: inf.uni-hamburg.de/en/inst/a…
is there a LaTeX template available for @lrec2020 submissions? i see M$ Word and OpenOffice on the conference web site, but doubt i could learn those tools in the remaining two-and-a-half weeks until the submission deadline.
Stephan Oepen retweeted
I don't know if you guys have heard whispers in the dark but another #ParsingTragedy😱 is coming up as part of #IWPT2020 This time it's about Deep Syntax, Enhanced UD and graphs. And no Bert allowed!
(just kidding🤣)
(Stay tuned for news from @danzem11, Gosse Bouma and I)
Stephan Oepen retweeted
Excited to see that the cross-framework (and multilingual) MR parsing shared task will run again next year at CoNLL2020
it is (almost) finished. schedule of presentations and links to system description papers for @conll2019 shared task on Cross-Framework Meaning Representation Parsing: mrp.nlpl.eu/index.php?page=6…
Stephan Oepen retweeted
Congratulations to LTG's @Murhaf on successfully defending his PhD dissertation today, with the title "Noun–Noun Compound Analysis: A Holistic Perspective"!
mn.uio.no/ifi/forskning/aktu…
Stephan Oepen retweeted
Vacant permanent position as associate professor in ML / #NLProc at the University of Oslo, Dept. of Informatics: jobbnorge.no/en/available-jo…
yes to more awareness for computational cost (and associated emissions), for example in our programme committees: technologyreview.com/f/61405…
Stephan Oepen retweeted
The slides of "Thirty Musts for Meaning Banking", presented today by Johan Bos at #dmr at #acl2019, are now available here: pmb.let.rug.nl/publications.…
materials from #ACL2019NLP tutorial on ‘graph-based meaning representations’ (with alexander koller and weiwei sun) are available on-line; even if all may seem said and done for now, all feedback will be much appreciated: github.com/cfmrp/tutorial
Stephan Oepen retweeted
Very interesting article about the recent research by @strubell et al on carbon footprint of NLP models, which also includes some opinions by yours truly. technologyreview.com/s/61363…
Stephan Oepen retweeted
Meaning representation toolkit now available for #MRP2019 @conll2019 shared task: github.com/cfmrp/mtool (evaluation period: July 8–22, 2019)
mtool (The Swiss Army Knife of Meaning Representation): scoring and support software for the @conll2019 Shared Task on Meaning Representation Parsing #MRP2019). pre-release on Microsoft GitHub: github.com/cfmrp/mtool
updated training data and premium-quality morpho-syntactic parses (thanks, @sebschu and @straka_milan) are now available to prospective participants in the @conll2019 shared task on Meaning Representation Parsing (#mrp2019): mrp.nlpl.eu
