Tweetovi
- Tweetovi, trenutna stranica.
- Tweetovi i odgovori
- Medijski sadržaj
Blokirali ste korisnika/cu @johnhewtt
Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @johnhewtt
-
Prikvačeni tweet
How do we design probes that give us insight into a representation? In
#emnlp2019 paper with@percyliang, our "control tasks" help us understand the capacity of a probe to make decisions unmotivated by the repr. paper: https://arxiv.org/abs/1909.03368 blog: https://nlp.stanford.edu/~johnhew/interpreting-probes.html …pic.twitter.com/1NA5hoyF7t
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
I'm giving a talk on designing and interpreting probing methods for understanding neural representations at EMNLP, Hall 2C, today at 1:30!
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
John Hewitt proslijedio/la je Tweet
Over the last few months, a group of recent NLP PhD applicants compiled some thoughts, perspectives, & advice on the application process. We're happy to share this blog post, and hope future applicants will find it helpful for years to come: https://blog.nelsonliu.me/2019/10/24/student-perspectives-on-applying-to-nlp-phd-programs/ … [1/3]
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Lots of hyperparameters when designing probes, and probing results conflate representation, probe, and data, making interpretration difficult. A control task can help design, and help interpret. code:https://github.com/john-hewitt/control-tasks …
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Selectivity can also help interpret probing results. Does ELMo1 have better part-of-speech representations than ELMo2? The accuracies suggest so, but probes can memorize -- and selectivity results show it's much easier to memorize from ELMo1.
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
We claim that good probes are "selective," achieving high accuracy on linguistic tasks, and low acc on control tasks. Between probes, small gains in linguistic acc can correspond to big selectivity losses; gains may be from added probe capacity, not repr properties.
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Our control tasks randomly partition the vocabulary, and force the probe to make the same output decision for words in the same subset. No linguistic structure, not reflective of repr, but learnable by the probe! Complex probes achieve high test accuracy on these tasks.
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
I'll be excitedly yammering about structural probes and finding syntax in unsupervised representations today at 4:15 in Nicollet B/C
#naacl2019. Even if you don't
parse trees, come by to learn a method to tell if your neural network softly encodes tree structures!Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
I enjoyed chatting with
@waleed_ammar and@nlpmattg on#nlphighlights about my paper with@chrmanning on finding syntax in word representations. I'm very grateful to have had this opportunity to talk (at length!) about my work!https://twitter.com/waleed_ammar/status/1125902568268320768 …Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
So a lot of people have arrived here; please read
@nsaphra's excellent take on neural net probes and@nelsonfliu's comprehensive neural net probing study, both also at#naacl2019 https://twitter.com/nsaphra/status/1099978662819057664 … Saphra: http://arxiv.org/abs/1811.00225 Liu: https://homes.cs.washington.edu/~nfliu/papers/liu+gardner+belinkov+peters+smith.naacl2019.pdf …
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
This claim, that parse trees are embedded through distances and norms on your word representation space, is a structural claim about the word representation space, like how vector offsets encode word analogies in word2vec/GloVE. We hope people have fun exploring this more! 4/4
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
These distances/norms reconstruct each tree, and are parametrized only by a single linear transformation. What does this mean? In BERT, ELMo, we find syntax trees approximately embedded as a global property of the transformed vector space. (But not in baselines!) 3/4
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Key idea: Vector spaces have distance metrics (L2); trees do too (# edges between words). Vector spaces have norms (L2); rooted trees do too (# edges between word and ROOT.) Our probe finds a vector distance/norm on word representations that matches all tree distances/norms 2/4
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Does my unsupervised neural network learn syntax? In new
#NAACL2019 paper with@chrmanning, our "structural probe" can show that your word representations embed entire parse trees. paper: https://nlp.stanford.edu/pubs/hewitt2019structural.pdf … blog: https://nlp.stanford.edu/~johnhew/structural-probe.html … code: https://github.com/john-hewitt/structural-probes/ … 1/4pic.twitter.com/G5cHK3kJ4w
Prikaži ovu nitHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Scott Aaronson's note https://www.scottaaronson.com/writings/bignumbers.html … is a delightful introduction to reasoning about large numbers, leading up to the Busy Beaver numbers. Years after finding that article, what fun to find Busy Beaver numbers in proofs on RNNs! https://arxiv.org/pdf/1711.05408.pdf …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Wondering under what circumstances visual signal is useful in translation? Feeling a desire for multimodal, multilingual NLP? Use our dataset of images representing words across 100 languages, and check out our poster in Session 3E with Daphne Ippolito http://aclweb.org/anthology/P18-1239 …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
We modeled derivational morphological transformations separately as orthographic and distributional functions, then combined: go see
@_danieldeutsch present our paper on English derivational morphology in oral session 6D today at ACL! http://aclweb.org/anthology/P18-1180 …Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
Learned a lot about LSTM behavior -- in very different ways -- from two excellent
@acl2018 papers: Sharp Nearby, Fuzzy Far Away... by@ukhndlwl, He He, Peng Qi, and@jurafsky, and LSTM as Dynamically Computed... by@omerlevy_ ,@kentonctlee,@nfitz,@lukezettlemoyer.Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
John Hewitt proslijedio/la je Tweet
Very excited to have my first paper at
@acl2018 with John Hewitt and Dan Roth: A Distributional and Orthographic Aggregation Model for Derivational MorphologyHvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi -
One of two first ACL papers! Daphne Ippolito and I spearheaded "Learning Translations via Images: A Large Multilingual Dataset and Comprehensive Study," to appear at
@acl2018! Happy to have worked with@brendandeecee, Reno Kriz, Derry Wijaya, and@ccb.Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.