Reconstructing Implicit Knowledge with Language Models.

Generating statements that explicate implicit knowledge connecting sentences in text.

aclweb.org/anthology/2021…

github.com/Heidelberg-NLP…
They make use of pre-trained language models which they refine by fine-tuning them on specifically prepared corpora that we enriched with implicit information and by constraining them with relevant concepts and connecting commonsense knowledge paths.
Manual and automatic eva. of the generations shows that by refining language models as proposed they can generate coherent & grammatically sound sentences that explicate implicit knowledge which connects sentence pairs in texts on both in-domain and out-of-domain test data
Spread the open source love!

If you know an amazing project drop me a message @philipvollet

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Philip Vollet

Philip Vollet Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @philipvollet

11 Jun
Insights from an open source influencer

I'm often asked how I get my content, over the years I've built an unusual technology stack for it

philipvollet.co

Some insights:
I use Feedly for most content inputs because I can access the content through a single API endpoint and scraping is often pure pain.

@feedly

Feedly saves me a lot of time and manual work.

feedly.com
To pull and enrich my GitHub content I use ghapi from @fastdotai which provides a 100% always-updated coverage of the entire GitHub API

ghapi.fast.ai
Read 8 tweets
11 Jun
XBNet: An Extremely Boosted Neural Network for Tabular Data with a novel architecture combining tree-based models with neural networks

arxiv.org/abs/2106.05239

github.com/tusharsarkar3/…
Trained by using a novel optimization technique, Boosted Gradient Descent for Tabular Data which increases its interpretability and performance.
This tweet is a collaboration with Tushar Sarkar so why not follow for first hand updates?

linkedin.com/in/tushar-sark…
Read 4 tweets
10 Jun
Deepface is a lightweight face recognition and facial attribute analysis framework in Python

$ pip install deepface

@serengil

Don't forget to spend some star love for the repository!

github.com/serengil/deepf…
It is a hybrid face recognition framework wrapping state-of-the-art models: VGG-Face, Google FaceNet, OpenFace, Facebook DeepFace, DeepID, ArcFace and Dlib

The library is mainly based on Keras & TensorFlow
This tweet is a collaboration with Sefik Ilkin Serengil so why not follow for first hand updates?

@serengil

Feel free to ask anything in the comments!
Read 4 tweets
10 Jun
Quant UX is a research, usability and prototyping tool to quickly test your designs & get data driven insights

@quant_ux

Quant-UX makes it simple to validate your ideas. Create a prototype, share a link & learn through user feedback and analytics

quant-ux.com
This tweet is a collaboration with Klaus Schaefers why not follow for first hand updates?

Feel free to ask anything!

linkedin.com/in/klaus-schae…
Read 4 tweets
9 Jun
Code for: Self-Attention Between Datapoints: Going Beyond Individual Input-Output Pairs in Deep Learning.

github.com/OATML/Non-Para… Image
Why not follow one of the authors Jannik Kossen for first hand updates?

@janundnik
Read 4 tweets
9 Jun
Figma-Low-Code use Figma designs directly in VUE.js applications.

The low code approach reduces drastically the hand-off time between designers and developers, reduces front-end code and ensures that the Figma design stays the single source of truth.

github.com/KlausSchaefers… Image
Don't forget to spend some star love for the repository!
This tweet is a collaboration with the author of Figma-Low-Code Klaus Schaefers why not follow for first hand updates?

Feel free to ask anything!

_

linkedin.com/in/klaus-schae…
Read 4 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!

:(