**👉🏼 2 A: Academia Challenges

👉🏼 2 B: Own projects**

<aside> 💡

Submissions must use the Apertus model family. For Track 2 this to means that submitted solutions must be built with Apertus. Other open-weights models can be used to support development, e.g as automatic judges during evaluation. Their role must be clearly described in the submission report.

💬 In case you have questions, join the conversation on Discord or send an email to “[email protected]”

💡 Check for available online intro sessions below where partners present their challenge and answer your questions.

</aside>


Academia Challenges + Own

Untitled

📅 Online Sessions Schedule

Date Time, CEST Partner Challenge/Topic Sign up Link
CHALLENGE INFO SESSIONS
02.10.2026 11:30 ZHAW See It, Say It, Pick It - Vision-Language Grounding for a Real Robot Arm https://luma.com/sxg60smd
02.10.2026 12:00 FHGR SmartStart: AI-Powered Job Interview Coach https://luma.com/g7mccjtk
02.10.2026 12:30 UZH SwissGov-RSD: Detecting Cross-Lingual Semantic Differences in Swiss Government Websites https://luma.com/zadclwjh
02.10.2026 13:30 OST Fact-Checking Political Claims with Official Vote Materials https://luma.com/qz7x62m8
02.10.2026 14:30 OpenParlData Extracting Parliamentary Affairs from PDFs into One Common Structure https://luma.com/mne4e0m5
CHALLENGE Q&A SESSIONS
08.10.2026 12:00 ZHAW See It, Say It, Pick It - Vision-Language Grounding for a Real Robot Arm https://luma.com/fmpw0ej9
08.10.2026 12:00 FHGR SmartStart: AI-Powered Job Interview Coach https://luma.com/dxtzhae0
08.10.2026 12:00 UZH SwissGov-RSD: Detecting Cross-Lingual Semantic Differences in Swiss Government Websites https://luma.com/18oeqo3d
08.10.2026 12:00 OST Fact-Checking Political Claims with Official Vote Materials https://luma.com/nkvf2m7y
08.10.2026 12:00 OpenParlData Extracting Parliamentary Affairs from PDFs into One Common Structure https://luma.com/3avq28dg