Florian

I'm interested in natural language processing and web technologies.

[Evan You @evanyou]

Excited to announce that @ViteConf is happening!
Get your free tickets :)

https://t.co/KzH5zvPRaf

[• nanou • @NanouuSymeon]

One website with everything you need 👇
https://t.co/LBKNSZUGrI

[Aurélien Geron @aureliengeron]

I just wrote a VS Code extension that removes the Python prompts (>>> and ...) and comments out the outputs. 😁 Useful when copy/pasting code from the shell.
It also goes the other way round if you prefer.
https://t.co/jWNWyGFsgT

[Mark Dalgleish @markdalgleish]

Just discovered that multiple people can screen share in Zoom at the same time if you're in dual monitor mode.

This is a game changer for remote pairing. Now we can split up and work in parallel while still being able to see what the other person is up to—just like in real life.

[Matthew Feickert @HEPfeickert]

I need to give some serious praise to fellow Scikit-HEP dev Hans Dembinski on his excellent monolens tool for interactive simulation of kinds of color blindness. https://t.co/qWHpIbIev7 It works really quite well and the fact that is is a #pipx install away is awesome!

[Rosamund Pearce @_rospearce]

The eastern sides of English cities tend to be poorer. This pattern can be traced back to the Industrial Revolution, when winds blew pollution east, causing the rich to flee west.

#dataviz https://t.co/8f5bLKzUoo

[elvis @omarsar0]

I'm excited about progress in GNNs and geometric deep learning in general because of its potential to address more complex problems.

@mmbronstein et al. recently published a 150+ pages book on Geometric Deep Learning. This is a must-read for ML students.

https://t.co/ysoRJp93lk

[Aran Komatsuzaki @arankomatsuzaki]

Finetuning Pretrained Transformers into RNNs

Successfully converts a pretrained transformer into its efficient linear-complexity recurrent counterpart with a learned feature map to improve the efficiency while retaining the accuracy.

https://t.co/yRfENT2ch2

[Shan He @heshan_cheri]

Check out https://t.co/UaqyF7gYZN! Built by the team behind https://t.co/SqbQgo2asx, https://t.co/1ckdGsUGia and h3 at @unfoldedinc. The ✨Unfolded Studio✨ is now in beta. We wrote about the idea behind it here -> https://t.co/MlMAaFokSp <-

[Xisen Jin @XisenJ]

(1/n) Can NLP models learn like humans within a single pass of a continuously evolving data stream? We explore this in our #emnlp2020 VisCOLL problem setup (with new datasets!) w/ @xiangrenNLP @ArkaSadhu29

Paper: https://t.co/NVxqdlaI1B
Project page: https://t.co/AvOIyv7ix4

[Jinhyuk Lee @leejnhk]

Our new work on learning accurate dense phrase representations for more than 60 billion phrases in the entire Wikipedia: https://t.co/7jL9scCu0k

+15-25% better than previous phrase retrieval models in Open QA without any sparse vectors, and you can use it as a dense KB, too!

[François Chollet @fchollet]

The Turing test was *never* a relevant goal for AI. We should remember that Turing never intended it as a literal test to be passed by a machine designed for that purpose, but as a philosophical device in an argument about the nature of thinking.

https://t.co/1M8hQaiPoi

[Alexis Ross @alexisjross]

Excited to share our preprint, "Explaining NLP Models via Minimal Contrastive Editing (MiCE)" 🐭 This is joint work with @anmarasovic and @mattthemathman

Link to paper: https://t.co/D612i7gQIm

Thread below 👇 1/6

[Papers with Code @paperswithcode]

⏪ Papers with Code: Year in Review. We’re ending the year by taking a look back at the top trending papers, libraries and benchmarks for 2020. Read on below!

https://t.co/jHoK4Cl94S

[Youngseog Chung @YoungseogC]

(1/6) Presenting: “Uncertainty Toolbox”, a python toolbox for predictive uncertainty quantification, calibration, metrics, and visualization.

URL: https://t.co/Envhu91gDo

[elvis @omarsar0]

I have read many books on NLP that cover topics to help obtain theory and how to build practical applications. From an educational standpoint, these are two NLP books, so far, that I found useful and can help an NLP student looking for hands-on experience and practical tips.

[Papers with Code @paperswithcode]

🎉 Papers with Code is expanding to more sciences! Today we launch new sites for physics, maths, CS, statistics and astronomy. You can use these sites to sync your code to show on arXiv. Explore our portal here: https://t.co/adn5t4rWDn

[Taro Makino @taromakino]

DNNs perform well on a range of medical diagnosis tasks, but do they diagnose similarly to humans?

In breast cancer screening, DNNs use different features than radiologists. Some are spurious, while others may represent new biomarkers.

https://t.co/kyMiLtSxw0
1/9

[Sasha Rush @srush_nlp]

These tutorial slides on "High Perf NLP" are really impressive. Every slide is current to the minute. Amazing set of diagrams.

https://t.co/o0o4SY6chR

(@gabriel_ilharco @Tim_Dettmers @IuliaTurc @kentonctlee Felipe Ferreira Cesar Ilharco)