🎉 Excited to share the recording of our workshop at the TUM.ai Summit last weekend 🎥 Our co-founder, muhtasham, gave an extensive and engaging overview of the workings of LLMs, from tokenization all the way to finetuning Llama2! ➡️youtu.be/Ug5c92gxDu0

📚 New Paper Release! 🧬 We conducted a thorough review and benchmarking of all proteasomal cleavage models available to date. We have also introduced some new architectures, one of which sets the new state-of-the-art for cleavage prediction! 🚀 biorxiv.org/content/10.110…

🗓️ Join us this Tuesday at 6pm CET, 9am PST, 12pm EST for a captivating talk by Amelie Schreiber, amelie_schreiber, about Protein Language Models! 🧪🧬 Amelie will explore the ESM-2 model family with its many variants and compare them to AlphaFold2. 🔗Find the link below 👇 (1/4)

🧪Did you miss or want to rewatch our captivating talk on Protein Language Models by amelie_schreiber? Good news! The event is now available on YouTube for you to rewatch and dive into the fascinating world of ESM-2 and its variants. 👉Check it out here: youtu.be/XAExm-G4QBc

CRAFT Your Dataset: Task-Specific Synthetic Dataset Generation Through Corpus Retrieval and Augmentation Presents a method for generating task-specific synthetic datasets using user-provided few-shot examples. 📝arxiv.org/abs/2409.02098 👨🏽‍💻github.com/ziegler-ingo/C…

[CL] CRAFT Your Dataset: Task-Specific Synthetic Dataset Generation Through Corpus Retrieval and Augmentation I Ziegler, A Köksal, D Elliott, H Schütze [University of Copenhagen & LMU Munich] (2024) arxiv.org/abs/2409.02098

Announcing our recent work “Multilingual Pretraining for Pixel Language Models”! We introduce PIXEL-M4, a pixel language model pretrained on four visually & linguistically diverse scripts: English, Hindi, Ukrainian & Simplified Chinese. #NLProc

Excited to share that our paper "Multilingual Pretraining for Pixel Language Models" has been accepted to the #EMNLP2025 main conference! Please see the thread below and the paper itself for more details.

1/2) I am very happy to finally share something I have been working on and off for the past year: "The Information Dynamics of Generative Diffusion" This paper connects the entropy production, divergence of vector fields and spontaneous symmetry breaking in a unified framework