Post-Doc at TU Munich, interested in just about anything in low-resource NLP
Joined October 2021
- Tweets32
- Following65
- Followers106
- Likes39
Thanks #BabyLM at #EMNLP2026 for the Challenge Awards for Qwenfant (aclanthology.org/2026.babylm…) and, on behalf of my student Serdar Gülbahar @cs_serdar, RecGPT (aclanthology.org/2026.babylm…)! It's always really fun and interesting working on the challenges for more efficient pretraining!
If you're a fan of my beautiful posters, I've got 2 new ones coming to #EMNLP2026:
1. Toppling the Hierarchy in Byte-level Language Modeling (arxiv.org/abs/2609.00463) at Poster Session D (Mon 11:00 - 12:30).
2. From Specialization to Generalization: Instruction-tuned LLMs for Robust Harmful Content Mitigation (arxiv.org/abs/2608.25605) at Poster Session F (Tue 9:00 - 10:30).
You can also see it Thu 14:15 - 15:30 at WOAH!
Great news! Our workshop what-llms-can-not-do.org is going to COLING 2027 in Macau! Hope to see you there!
Ever feel like it's too hard to keep track of what LLMs cannot do as well as humans? We're making your life easier over at: what-llms-can-not-do.github.…
We're compiling a list of papers testing the abilities of LLMs against humans. Check it out! And you can help contribute too!
Lukas Edman retweeted
Excited to kickstart this effort with @glnmario and a star-studded advisory team! 🚀
Also, we're hiring for 2 MTS and 1 Ops manager! Come build the future of white-box audits with us: parallx.ai/careers
Introducing Parallax 🔍
We're a new UK/EU-based nonprofit research lab building scalable methods and infrastructure to audit the beliefs, goals, and plans that shape model behaviours in long-horizon agentic evaluations.
parallx.ai — Thread 🧵 1/
Lukas Edman retweeted
Machine translation is not solved and it will take a while for it to be done arxiv.org/abs/2609.04173
Everyone talks about how great ChatGPT and Claude are, but Olmo is literally in the world cup finals #NLProc
Lukas Edman retweeted
🍼 Mask and You Shall Receive: Optimizing Masked Language Modeling for Pre-training BabyLMs
👶 @LukasNLP, Alexander Fraser
arxiv.org/abs/2510.20475
Thanks to everyone who showed up at my posters: EXECUTE, and Positional Overload w/ Lukas Kinder! #ACL2025 #ACL2025NLP
Happy to announce two papers at #ACL2025!
1. We made an extension to our CUTE benchmark, EXECUTE! We find that LLMs actually *do* have the ability to do character-level manipulation, but language bias, and tokenization too, get in the way. openreview.net/forum?id=m955…
1/2
Thanks to everyone who stopped by my poster at #EMNLP2024 today! Had some really fun and interesting discussions with you all :)
I'm happy to announce that our LLM benchmark for measuring orthographic understanding, CUTE, was accepted to EMNLP main! Looking forward to seeing you all in Miami! #NLProc
Arxiv: arxiv.org/abs/2409.15452
HF Leaderboard: hf.co/spaces/leukas/cute_lea…
Github: github.com/leukas/cute
Lukas Edman retweeted
Very excited to present my work at @CisLmu tomorrow; thanks for having me! 😄
🎓 We're thrilled to host Gabriele Sarti (@gsarti_) in our PhD seminar series tomorrow, July 16th, from 12:00-13:00 in Oe67 BU 101! Join us for his talk on interpreting context usage in generative language models, featuring the Inseq toolkit and PECoRe framework. 🕛Don't miss it!
In case you missed it, you can check out the paper and models at github.com/Leukas/charlevelm…, and feel free to reach out!
I'm happy to announce that my work "Are Character-level Translations Worth the Wait? Comparing ByT5 and mT5 for Machine Translation" with @gsarti_, @_atoral, @gertjanvannoor1, and @AriannaBisazza was accepted to #TACL! I'll be presenting it at #EACL2024!
arxiv.org/abs/2302.14220
Lukas Edman retweeted
After a year of restless development, I'm finally happy to announce Inseq, a new tool to democratize post-hoc interpretability of sequence generation models 🐛 github.com/inseq-team/inseq #nlproc #xai
Some highlights 👇 1/
