Laura Skylaki

ML Applied Research, Deep Learning, NLP, Bioinformatics.

[Yann LeCun @ylecun]

I can't make it to #CVPR2023 this year, but @MetaAI / FAIR has a huge presence, as usual.

More info: https://t.co/9d3ZJ8QMlv

[Laura Skylaki @lauraskylaki]

The mystery of a well targeted, well timed ad, frankly quite creepy, the targeting, not the ad

[Ilias Chalkidis @KiddoThe2B]

Our paper "LEGAL-BERT: The Muppets straight out of Law School" with @ManosFergas, @NeuRulller, @nikaletras and @ionandrou, has been accepted in Findings of #EMNLP2020. Arxiv pre-print available at: https://t.co/hCRAsNCw4V. Brief summary: 1/5

[JohnSnowLabs @JohnSnowLabs]

Brian Ulicny @bulicny from @thomsonreuters presents different summaries of legal cases - which was generated by a human and which by machine learning? See the impressive results of what #NLP can do with legal cases https://t.co/fv0RjPIOcd #nlpsummit

[Laura Skylaki @lauraskylaki]

Check it out for state-of-the-art summarisation insights!

[ODSC (Open Data Science Conference) AI @_odsc]

#ODSCWest Virtual Conference Speakers. https://t.co/4YHxQODB6V #DeepLearning #NLP @NHristozova @thomsonreuters

[Laura Skylaki @lauraskylaki]

Article of the year: what's the deal with birds?

[Dan Baldassarre @evornithology]

Been getting a lot of emails from this predatory journal lately, so I decided to submit this manuscript. Stay tuned. 🤞🤞🤞 https://t.co/nv1f2PausF

[Laura Skylaki @lauraskylaki]

The conference call bingo. The "sound of someone typing, possibly with a hammer" made me laugh.

[Kristin Politi @KristinPoliti]

@Toaster_Pastry This conference call bingo is making the rounds

[Sebastian Ruder @seb_ruder]

This is a *really* extensive repo containing ~380 BERT-related papers sorted into downstream tasks, modifications, probes, multilingual models, and more. Nice job, @stomohide!

https://t.co/CaY0nZxDhV

[500WS Zürich @500WSZurich]

Our #Wikipedia #editathon event is fast approaching! Join us now and help us create Wikipedia pages for notable #WomenInSTEM. @WikimediaCH will guide us through all the steps, so all you need to do is register https://t.co/zosn71wtsY and nominate someone https://t.co/UWg5M2rPjh

[Laura Skylaki @lauraskylaki]

Today I'll be talking about our recent advances in AI applications for the legal domain at the Swiss Legal Tech Association @swisslegaltech. See you there! #workingattr #trlabs

[SLTA @swisslegaltech]

The Swiss Legal Tech Association is happy to present current solutions for the legal sector which are utilizing artificial intelligence.

Save your seat on time: https://t.co/jKMgVFzT3n

#legaltech #law #innovation #artificialintelligence

[Laura Skylaki @lauraskylaki]

I loved this article by Kate Shram from our careers blog. More than enough! #workingattr #BeYouAtTR #womeninscience #womenintechnolodgy
https://t.co/Us4iEMOS1O

[Thomson Reuters @thomsonreuters]

For the 8th consecutive year, we're proud to have earned 100% on the Corporate Equality Index, the @HRC's annual scorecard assessing #LGBTQ workplace #equality.

Read more about or commitment to the LGBTQ community in the US and worldwide here:

https://t.co/6S9ICte3GR

[Uwais Iqbal @_uwaisiqbal]

Last month I had the opportunity to attend the #odisummit 2019 with colleagues from #TRLabs - I've put together some reflections and takeaways from the summit - Enjoy!
#opendata #workingattr #dataforgood #innovation

[Viki Male @VikiLovesFACS]

At conferences, I use Inigo Montoya’s guide to Networking Success (Yes, really. Otherwise I always forget to say my name🤦🏻‍♀️)

1. Greeting
2. Name
3. Relevant personal connection
4. What to expect from the encounter

#BSI2019

[Laura Skylaki @lauraskylaki]

@ilyasut Oh no they didn't! 😂

[Laura Skylaki @lauraskylaki]

“You are precise,” she continued, which I initially construed as a compliment intended to interrupt the flow of criticism—until she elaborated. “So when people with common sense get frustrated, you prevail.”

Yeah! Take that, people with common sense.

https://t.co/dfVmXnltrl

[Laura Skylaki @lauraskylaki]

Great to read, but really edited in the sense that they would pick the most coherent answer out of five attempts
https://t.co/PIvWQp6ipG

[Laura Skylaki @lauraskylaki]

Great work, bookmarked.

[Quoc Le @quocleix]

Want to improve accuracy and robustness of your model? Use unlabeled data!

Our new work uses self-training on unlabeled data to achieve 87.4% top-1 on ImageNet, 1% better than SOTA. Huge gains are seen on harder benchmarks (ImageNet-A, C and P).

Link: https://t.co/ZYDaef6sdp

[Scientific American @sciam]

Why do all the pictures you take underwater look blandly blue-green? The answer has to do with how light travels through water. @dakkaynak, an oceangoing engineer, has figured out a way to recover the colorful brilliance of the deep.

Read the full story: https://t.co/0hgKoPkS4B

[Last Week in AI @Last_Week_in_AI]

Check out our latest effort to combat AI hype!

Even well intentioned AI media coverage can get things wrong in a bunch of subtle ways.

So we surveyed AI researchers, and summarized the things they say to watch out for.

Let us know what you think!

https://t.co/OCkKowf6zr