Might be interesting for many of you, so save the date... It's a hybrid talk, so you may just follow online.
Best, Timo
-------- Weitergeleitete Nachricht -------- Betreff: [sira-news] Invitation: SIRA Brown Bag Talk by Martin Jaggi on Fri November 21, 2025 13:15-14:15 at EPFL and online Datum: Wed, 29 Oct 2025 13:36:07 +0100 Von: Martin Glinz glinz@ifi.uzh.ch An: sira-news@list.inf.unibe.ch Kopie (CC): sira-board@list.inf.unibe.ch
Dear SIRA members,
You are cordially invited to our next SIRA Brown Bag Talk which will take place as a part of the 2025 Annual SIRA Conference.
Date and time: Friday November 21, 2025 13:15-14:15
Speaker: Martin Jaggi (EPFL) Title: Apertus: Democratizing Open and Compliant LLMs for Global Language Environments
Venue: - Physically at EPFL, BC Building, room BC 420 and - Online at https://uzh.zoom.us/j/66608357823?pwd=cNAvZKibfpI06bb4zpgMKtREbngvhZ.1
Abstract: We present Apertus, a fully open suite of large language models (LLMs) designed to address two systemic shortcomings in today's open model ecosystem: data compliance and multilingual representation. Unlike many prior models that release weights without reproducible data pipelines or regard for content-owner rights, Apertus models are pretrained exclusively on openly available data, retroactively respecting robots.txt URL exclusions and filtering for non-permissive, toxic, and personally identifiable content. To mitigate risks of memorization, we adopt the Goldfish objective during pretraining, strongly suppressing verbatim recall of data while retaining downstream task performance. The Apertus models also expand multilingual coverage, training on 15T tokens from over 1800 languages, with ~40% of pretraining data allocated to non-English content. Released at 8B and 70B scales, Apertus approaches state-of-the-art results among fully open models on multilingual benchmarks, rivalling or surpassing open-weight counterparts. Beyond model weights, we release all scientific artifacts from our development cycle with a permissive license, including data preparation scripts, checkpoints, evaluation suites, and training code, enabling transparent audit and extension. Link: https://www.swiss-ai.org/apertus
Bio: MARTIN JAGGI is an associate professor at EPFL, where he is leading the Machine Learning and Optimization Laboratory. Before that, he was a post-doctoral researcher at ETH Zurich, at the Simons Institute in Berkeley, and at École Polytechnique in Paris. He has earned his PhD in Machine Learning and Optimization from ETH Zurich in 2011, and a MSc in Mathematics also from ETH Zurich.
Please forward this invitation also to the members of your group. You find this announcement also at https://sira.swissinformatics.org.
Please note that the 2025 Annual SIRA Conference meeting will take place right after Martin Jaggi’s talk (same place, also physically and online). The formal invitation has been sent by e-mail to all SIRA representatives. SIRA members who are neither a SIRA representative nor an official substitute for a SIRA representative are invited to attend, but have no vote.
Best regards
Martin Glinz SIRA President