rayane ghilene

NLP Engineer and Researcher at CYU | Electrical and Computer Engineering at ENSEA | Research masters at CYU | Part-time Musician

[rayane ghilene @rayenghilen]

@jerbosmans This is next level Jerry.. Bravo

[rayane ghilene @rayenghilen]

@TheAppleDesign *Charger not included in the box

[EMNLP 2026 @emnlpmeeting]

📢 The First Call for Papers for EMNLP 2026 is officially out! 📝
We welcome long & short papers featuring original research on empirical methods for NLP.

🗓️ ARR Submission Deadline: May 25, 2026
🔗 Read the full CFP here: https://t.co/GU2kaISjUG

#EMNLP2026

[Proton Drive @ProtonDrive]

DON'T UPLOAD YOUR FAMILY PHOTOS TO AI
DON'T UPLOAD YOUR FAMILY PHOTOS TO AI
DON'T UPLOAD YOUR FAMILY PHOTOS TO AI
DON'T UPLOAD YOUR FAMILY PHOTOS TO AI
DON'T UPLOAD YOUR FAMILY PHOTOS TO AI

[rayane ghilene @rayenghilen]

@Oxdave

[rayane ghilene @rayenghilen]

@karthikponna19

[rayane ghilene @rayenghilen]

@Wisdom_HQ

[rayane ghilene @rayenghilen]

@Parisianaes1

[rayane ghilene @rayenghilen]

👨‍💻 Code: https://t.co/29r0NdtoZY

[rayane ghilene @rayenghilen]

You can finally check out our paper, now out in the Findings of hashtag #EMNLP2025

📄Paper link: https://t.co/4k2tocTwdX

We explore how artifact-guided pretraining can make masked language models more interpretable and better at detecting socially unacceptable discourse across diverse datasets. Our approach combines linguistic token importance scoring with selective masking to improve explainability and generalization.

Grateful to my co-authors Dimitra Niaouri, Michele Linardi, and Julien Longhi, and to the ARENAS project for the support.

[rayane ghilene @rayenghilen]

@virby_ai VHDL 👑

[rayane ghilene @rayenghilen]

@luquinhasedit @aftermagics old pepsi logo

[rayane ghilene @rayenghilen]

@aftermagics leave illustrator out of this

[rayane ghilene @rayenghilen]

@karrisaarinen Private relay is quite handy

[rayane ghilene @rayenghilen]

Maturity is accepting Inter’s default letter spacing.

[Rohan Paul @rohanpaul_ai]

The new steam age.

This is actually becoming true in many cases.

It's possible to do so much more on your own now.

[rayane ghilene @rayenghilen]

@sama we want to pin chats in the history section. Thanks 🙏

[Matthew Berman @MatthewBerman]

My new employee revealed he used AI to negotiate his salary with me.

I revealed I used AI to negotiate back...

[rayane ghilene @rayenghilen]

@ZacksJerryRig I think the whole internet is waiting on you

[rayane ghilene @rayenghilen]

Thank you for reporting this to the team. It would be great to include a feedback button, as that could really help with the model alignment.

It would also be great to have a PII (Personally Identifiable Information) Masking feature to keep users safe when communicating with chatbots.

BTW, I have the necessary skills to solve these problems. Happy to help if you are hiring :)

[rayane ghilene @rayenghilen]

Not really @Grammarly, but Thanks.