Twitter | Search | |
Search Refresh
Rahul Goel Jun 16
Using Language Modeling as additional supervision before fine tuning on task at hand = Self Supervised Learning for NLP
Reply Retweet Like
Emiel van Miltenburg Jun 17
people: If you need to annotate short sentences, and you don’t want to install any new tools, this might be useful: This GitHub gist shows how to create an Excel sheet where people can annotate sentences at the word level, using dropdown menus.
Reply Retweet Like
Hugging Face Jun 20
RT if you want XLNet in
Reply Retweet Like
Suzan Verberne 🤹‍♀️ 2h
Stacked word embeddings for named entity recognition
Reply Retweet Like
Yuval Marton Jun 19
This metric has been with us since 1990. It was the first shot in the statistical revolution during which Royal Rules were guillotined. It deserves more respect! It deserves its own shop! I give you: The Bleu!
Reply Retweet Like
Abusive Language WS 20h
Hey folks! Awesome news: We’ll be able to support a few student’s participation in the workshop so if you (or someone you know) is working on online abuse and want to join please have them fill out this form:
Reply Retweet Like
Wluper May 29
Reply Retweet Like
Amit Moryossef Jun 21
Why am I getting this 2 years old call for paper now?
Reply Retweet Like
Accenture Federal 15h
Cloud-based services & APIs make it easier to get started with .
Reply Retweet Like
Prasanth K Jun 20
LDC is one of my less-that-fav orgs assoc. with . I know how & why it was setup the way it is, but it *used to* and *still* reminds me of a wall for disenfranchised researchers which I've been at various times in the last 10 years.
Reply Retweet Like
ictclas.nlpir Jun 20
We introduce a new type of deep contextualized word representation that models both (1) complex characteristics of word use (e.g., syntax and semantics), and (2) how these uses vary across linguistic contexts (i.e., to model polysemy).
Reply Retweet Like
GraphAware 4h
From Boston, DC, Helsinki, Melbourne, London, Prague and many more cities, developers and clients are gathering in Lecce, Italy, for the first hackathon !
Reply Retweet Like
Daniel Hoadley 2h
The Blackstone model we are building at is really coming along. The model is getting really good at spotting case names in context and has generalised well enough to find unconventional and overseas citation patterns.
Reply Retweet Like
Ari Bornstein 36m
My presentation on Beyond Word Embeddings from PyCon Israel 2019 strongly recommend watching the video in 1.5x speed.
Reply Retweet Like
Tiago Torrent 19h
, , and people, Global FrameNet is here! Come say hi!
Reply Retweet Like
Kilian Gebhardt Jun 20
I hate it, if there are some obvious and in theory easy tweaks you can do to your model, but you cannot do it (at all or without investing countless hours), because the original software was never released or the data is not available.
Reply Retweet Like
Omnia Zayed 2h
Excited/delighted to start the summer schools in corpus linguistics at Lancaster University... let the adventure begin ;) ^^
Reply Retweet Like
Marek Rei Jun 17
A great talk on how to use BERT for knowledge base construction, by at the inaugural seminar.
Reply Retweet Like
Graham Neubig Jun 19
lazyweb: is there any easy-to-use toolkit for extracting parallel sentences from Wikipedia? I know these have existed at some point, but I don't know good modern ones.
Reply Retweet Like
Yuval Marton Jun 19
If Twitter doesn’t crop or downgrade the resolution to nothing, look at the three street signs. The middle one is blurry, but with the right language model your can figure it out
Reply Retweet Like