The below is an off-site archive of all tweets posted by @texttheater ever

February 2nd, 2018

Was mich bei Google Maps schon seit etlichen Neugestaltungen richtig nervt, ist seine extreme Zurückhhaltung, was das Anzeigen von Straßennamen betrifft.

via Twitter Web Client

skeptikantin Ich glaub, hier ist man dagegen. Also dafür gegen das Dafür beim Dagegen.

HAB ICH ZU VIEL VERSPROCHEN HAB ICH GEFRAGT.

pic.twitter.com/AX5dCaPbHw

via Twitter for Android (retweeted on 5:12 PM, Feb 2nd, 2018 via Twitter Web Client)

kathrinpassig There's probably no algorithmic transparency. Now stop worrying and enjoy your life. eurozine.com/black-box-stat… (English translation of my last @redaktionmerkur
essay, thanks to @Eurozine.)

via TweetDeck (retweeted on 5:04 PM, Feb 2nd, 2018 via Twitter for Android)

Autorwesen Maybe the real treasure was the psychological trauma we got along the way.

via Twitter for iPhone (retweeted on 3:20 PM, Feb 2nd, 2018 via Twitter Web Client)

nsaphra A neat exploration from @boknilev et al. of how semantic and syntactic information is built from lower layers and new contextual information within neural mt models, building on their 2017 implicit morphology results. arxiv.org/pdf/1801.07772…

via Twitter Web Client (retweeted on 3:19 PM, Feb 2nd, 2018 via Twitter Web Client)

egrefen Delighted that our paper on neural nets and logic was accepted to ICLR'18.
openreview.net/pdf?id=SkZxCk-…
Summary: on highly and heterogeneously structured tasks (detecting logical entailment), models are ranked: semantically-aware > syntactically-aware > LSTM > ConvNet.

via Twitter for iPhone (retweeted on 1:36 PM, Feb 2nd, 2018 via Twitter Web Client)

Phylan ME: *makes typo while entering a number*

EXCEL: WAS THAT A DATE

ME: no I meant t-

EXCEL: THAT WAS A FUCKING DATE

ME: it doesn't even make sen-

EXCEL: MAY 12TH 1382. LOOK I EVEN FORMATTED IT. IT IS THIS FOREVER

via Twitter Web Client (retweeted on 9:40 AM, Feb 2nd, 2018 via Twitter Web Client)

LeonDerczynski The Broad Twitter Corpus v1.0 is released! bit.ly/2nzLxMZ - 150K tokens of tweets spread over time, place, and social use. This corpus is also a @github repo, so please send pull requests and create issues. Paper: bit.ly/2nTM3aF pic.twitter.com/6V0jWIV5fm

via TweetDeck (retweeted on 8:03 AM, Feb 2nd, 2018 via Twitter for Android)