𝙃𝙚𝙡𝙡𝙤 𝙒𝙤𝙧𝙡𝙙! 🚀 We, the AI department at Fraunhofer HHI, are excited to launch our X channel! We’ll be sharing news, research, and events about Explainability, Efficiency, and Applications of AI. Follow us to stay updated and join the AI Community! 🚀 #AI #Research #HHI

🎉 10 Years of AI at Fraunhofer HHI 🎉 We celebrated the 10-year anniversary of our AI department - a decade of innovation and research breakthroughs. Thanks to everyone who's been part of this journey. Here's to many more successful years ahead! #Fraunhofer #AI #anniversary

I'am thrilled to announce that our paper "Generative Fractional Diffusion Models (GFDM)" has been accepted at #NeurIPS2024! 🚀 📜Check-out our paper: arxiv.org/abs/2310.17638 🌟Star our repository for code release: github.com/GabrielNobis/g… 🧵1/6

Building trust in AI! 🔒 #TrustworthyAI Prof. Wojciech Samek co-hosted the AI Forum – a unique event bringing together research, industry and regulators to make AI safe, reliable and sustainable. Thanks to TÜV-Verband, BSI and all participants for their contributions! 🙏

Let's go, #NeurIPS2024! 🚀 Join us today from 11am-2pm at our poster session in East Exhibit Hall A-C, #2700. We’re presenting Generative Fractional Diffusion Models: 📜Check-out our paper: arxiv.org/abs/2310.17638 💻Check-out our code: github.com/GabrielNobis/g…

Congrats to Fraunhofer HHI #XAI experts Wojciech Samek & Sebastian Lapuschkin! @Handelsblatt has listed them among the most cited #AI researchers in Germany, and Fraunhofer HHI as one of the most important #AIhubs in the country. 👉hhi.fraunhofer.de/kiforschungfra…

🎄 Season’s Greetings from the Fraunhofer AI Group! 🎄 Thank you for an amazing year of projects and collaboration. We truly appreciate your support. Wishing you a joyful and relaxing holiday season! 🌟

✨ Happy New Year from the Fraunhofer AI Group! ✨ Thank you for making this year unforgettable with your inspiring work and collaboration. We look forward to more achievements ahead. Wishing you a year full of new opportunities and happiness! 🥂 #FraunhoferHHI #FraunhoferAI

We celebrated Lunar New Year 🪵🐍 by welcoming a delegation from 🇸🇬 Singapore to our Science Tech Space @ HHI showroom! Thank you, Deputy Prime Minister Heng swee keat, Ambassador ChongHock LEE (MFAsg), and your delegation for visiting us! 🙏

Representing Fraunhofer HHI at the 🇪🇺Presidency Workshop “Understanding Communication in Civil Protection”, our expert Katharina Weitz emphasized: For new technology to succeed in civil protection, involving all key stakeholders from day one isn’t just important—it’s essential!⛑

Fraunhofer HHI researchers and international experts published a new paper in Nature Geoscience 🌍 To promote the use of #XAI in the #Geosciences, they surveyed researchers and offer 4 actionable recommendations to advance its use. 👉 nature.com/articles/s4156…

Comment: Explainability can foster trust in the use of AI in geoscience ECMWF AI Fraunhofer HHI nature.com/articles/s4156…

🚨 New paper alert! 🚨 We’re excited to share our latest work on interpretability evaluation: "Evaluating Interpretable Methods via Geometric Alignment of Functional Distortions" 📜 Accepted at TMLR 🎉 🔥 Survey certification 🔥 📖 Read: openreview.net/pdf?id=ukLxqA8…

🎥 In a new feature by DW Deutsche Welle, Fraunhofer HHI expert Monique Kuglitsch explains why global standards are the game-changer for AI in disaster resilience. 🖱️ Skip to 4:36 to get Monique’s insights: dw.com/de/mit-tech-un… 🔗More information: hhi.fraunhofer.de/en/news/nachri…

🎉 On #GirlsDay 2025, Fraunhofer HHI welcomed schoolgirls to explore STEM careers through hands-on workshops. 🤖 In the AI session, led by Dr. Katharina Weitz they trained models, tackled bias & tested explainable AI tools. 💡 A day full of learning, tech & inspiration!

🚀 New demo! Explore CLIP’s hidden concepts with SemanticLens. Integrated 16 SAEs from ViT Prisma (github.com/soniajoseph/Vi…) 💻: semanticlens.hhi-research-insights.eu 📎: arxiv.org/pdf/2501.05398 Fraunhofer HHI TU Berlin BIFOLD #ViTPrisma #SemanticLens #SAE #XAI #Interpretability

This winter, I joined Egor Zverev @ACL2025 in exploring how to make certain LLM inputs non-executable, in order to defend LLMs from prompt injections. The results are both promising and conceptually interesting. Find me this week at ICLR to discuss!

ICL allows LLMs to adapt to new tasks and at the same time enables them to access external knowledge through RAG. How does the latter work? TL;DR we find that certain attention heads perform various, distinct operations on the input prompt for QA! arxiv.org/abs/2505.15807 1/

New paper from our team on making LLM citations more transparent. Check it out: arxiv.org/abs/2505.15807

🔬 Happy Research Appreciation Day! On July 5, we celebrate the people behind the progress — our researchers. 🌍 Check out our posts to see what our team at Fraunhofer HHI values most about research in AI and XAI. #ResearchAppreciationDay #AI #XAI #FutureOfAI