Feliçia Körner

fkoerner.bsky.social

22 September 2026 18:57

I'm so glad you looked into this! Using logit lens to identify a model's latent language always felt a bit incomplete to me

Aaron Mueller

amuuueller.bsky.social

08 September 2026 12:24

Most interpretability is post hoc. How can we understand when a feature is learned during training? We apply crosscoders to understand when models learn syntactic features, and when multilinguality arises. Chat with @bayazitdeniz.bsky.social at ACL! bsky.app/profile/baya...

Badr AlKhamissi

bkhmsi.bsky.social

20 October 2025 12:15

🚀 Excited to share a major update to our “Mixture of Cognitive Reasoners” (MiCRo) paper! We ask: What benefits can we unlock by designing language models whose inner structure mirrors the brain’s functional specialization? More below 🧠👇 cognitive-reasoners.epfl.ch

Badr AlKhamissi

bkhmsi.bsky.social

25 September 2025 16:33

Now that the ICLR deadline is behind us, happy to share that From Language to Cognition has been accepted as an Oral at #EMNLP2025! 🎉 Looking forward to seeing many of you in Suzhou 🇨🇳