Hosein Mohebbi's Avatar

Hosein Mohebbi

@hmohebbi.bsky.social

PhD candidate at Tilburg University, doing research on interpretability for text and speech. https://hmohebbi.github.io/

107 Followers  |  109 Following  |  5 Posts  |  Joined: 19.11.2024  |  1.8638

Latest posts by hmohebbi.bsky.social on Bluesky

Interspeech paper title: What do self-supervised speech models know about Dutch? Analyzing advantages of language-specific pre-training

Authors: Marianne de Heer Kloots, Hosein Mohebbi, Charlotte Pouw, Gaofei Shen, Willem Zuidema, Martijn Bentum

Interspeech paper title: What do self-supervised speech models know about Dutch? Analyzing advantages of language-specific pre-training Authors: Marianne de Heer Kloots, Hosein Mohebbi, Charlotte Pouw, Gaofei Shen, Willem Zuidema, Martijn Bentum

✨ Do self-supervised speech models learn to encode language-specific linguistic features from their training data, or only more language-general acoustic correlates?

At #Interspeech2025 we presented our new Wav2Vec2-NL model and SSL-NL evaluation dataset to test this!

πŸ“„ arxiv.org/abs/2506.00981

⬇️

27.08.2025 14:31 β€” πŸ‘ 19    πŸ” 6    πŸ’¬ 1    πŸ“Œ 0

Had such a great time presenting our tutorial on Interpretability Techniques for Speech Models at #Interspeech2025! πŸ”

For anyone looking for an introduction to the topic, we've now uploaded all materials to the website: interpretingdl.github.io/speech-inter...

19.08.2025 21:23 β€” πŸ‘ 40    πŸ” 14    πŸ’¬ 2    πŸ“Œ 1
Posthoc Disentanglement of Textual and Acoustic Features in Self-Supervised Speech Encoders | AIW'25
YouTube video by Hosein Mohebbi Posthoc Disentanglement of Textual and Acoustic Features in Self-Supervised Speech Encoders | AIW'25

Sadly, I couldn't make #ICML2025 in person to present our work at the exciting Actionable Interpretability Workshop @actinterp.bsky.social.
πŸ“ΊThe recording is online: www.youtube.com/watch?v=QkDF...
+ Poster (on the workshop's website)
πŸ‘‹Check it out and feel free to get in touch for any questions!

17.07.2025 13:52 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

πŸ“’ In our #Interspeech2025 tutorial, we'll provide a structured overview of interpretability methods, their applications, challenges, and limitations when applied to spoken language models.
βž• hands-on experience
interspeech2025.org/tutorials

13.06.2025 13:58 β€” πŸ‘ 5    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

Have you already found #BlackboxNLP on Bluesky?πŸŽ‰

15.05.2025 18:41 β€” πŸ‘ 5    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0
Post image

πŸŽ‰ Our Actionable Interpretability workshop has been accepted to #ICML2025! πŸŽ‰
> Follow @actinterp.bsky.social
> Website actionable-interpretability.github.io

@talhaklay.bsky.social @anja.re @mariusmosbach.bsky.social @sarah-nlp.bsky.social @iftenney.bsky.social

Paper submission deadline: May 9th!

31.03.2025 16:59 β€” πŸ‘ 43    πŸ” 16    πŸ’¬ 3    πŸ“Œ 3
Post image

My PhD thesis is now available online! It is called "Finding Structure in Language Models" and presents the work I have done in the past 4 years on the intersection of NLP, interpretability, and linguistics.

I'll defend it on December 10, streamed here: jumelet.ai/phd

arxiv.org/abs/2411.16433

🧡

29.11.2024 15:49 β€” πŸ‘ 115    πŸ” 14    πŸ’¬ 8    πŸ“Œ 3

The sec. background in (aclanthology.org/2023.emnlp-m...) explains how representations are made in the encoder/decoder of speech transformers.

27.11.2024 08:02 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

πŸ™‹πŸ»β€β™‚οΈ

23.11.2024 23:16 β€” πŸ‘ 1    πŸ” 0    πŸ’¬ 0    πŸ“Œ 0

@hmohebbi is following 20 prominent accounts