TANGO Project

From all TANGO Project partners, we wish you a very happy and safe holiday season! We look forward to sharing updates about the project and upcoming events in 2025.

🚀 New TANGO Project Blogpost! 🌐 Accelerating Trustworthy AI with POEM in Time-Sensitive Scenarios In critical fields like autonomous driving, real-time medical diagnostics, and high-frequency trading, machine learning (ML) models must not only deliver accurate results but also offer instant and trustworthy explanations. Enter POEM (Prefetched Offline Explanation Model) — a ground-breaking framework developed by Istituto di scienza e tecnologie dell'informazione "A. Faedo" (Cnr-Isti) and the Università di Pisa to address these challenges. 💡 What Makes POEM Special? Unlike traditional methods like LIME or SHAP, POEM is designed specifically for real-time applications, achieving unparalleled speed and reliability through a novel dual-phase approach: 1️⃣ Offline Phase: Precomputes detailed explanations (exemplars, counter exemplars, saliency maps) for representative data points, storing them in an explanation base. 2️⃣ Online Phase: Rapidly retrieves the most relevant precomputed explanation for new instances, ensuring decisions are accompanied by immediate, high-quality insights. 📊 The Results Speak for Themselves: Speed: Execution time reductions of up to 96.5% compared to previous models like ABELE. Quality: Exemplars and saliency maps provide semantically rich, actionable insights, validated across datasets like MNIST and FASHION. Versatility: Modular design supports adaptation to other data types, paving the way for explainable AI in diverse domains. 🌟 Why It Matters: POEM sets a new standard for explainable AI by balancing transparency, efficiency, and accuracy, ensuring AI remains a trusted partner in high-stakes, time-sensitive environments. 🔗 Dive into the full story here: https://lnkd.in/eD3eeM5z Let’s advance AI that earns trust and delivers impact where it matters most. #AI #ExplainableAI #POEM #TrustworthyAI #MachineLearning #Innovation #RealTimeAI

🚀 Exciting News! 🚀 We're thrilled to announce the 2nd Adaptive XAI (AXAI) Workshop, to be held in conjunction with ACM IUI 2025 in Cagliari, Sardinia on March 24, 2025! 🌍 As AI becomes more deeply embedded in daily decision-making, the need for clear, adaptive communication between humans and AI systems is more crucial than ever. The AXAI 2025 Workshop focuses on designing human-centric interfaces that adaptively explain AI decision-making in ways that resonate with diverse users and contexts. 🔑 Important Dates: Submission Deadline: January 13, 2025 Notification: January 20, 2025 Workshop Date: March 24, 2025 💡 Topics of Interest include: • Personalized XAI interfaces • Novel interaction modalities (including immersive and multimodal experiences) • Ethical considerations in adaptive XAI • Domain-specific applications (e.g., healthcare, finance) Evaluation methodologies and much more! 📑 More information at: https://axai.trx.li/ For submission guidelines and further details, check out the workshop page: https://lnkd.in/g4Nu-bDM We invite researchers, practitioners, and AI enthusiasts to join us in shaping the future of human-AI interaction! Feel free to share with your network! 🌟 #AI #XAI #MachineLearning #HumanAIInteraction #ACMIUI2025 #ResearchWorkshop #AdaptiveSystems #AIethics #TechnologyInnovation

💥🔤 TANGO ABC marches on! L is for... #HumanInTheLoop The #TANGO_Project, supported by €7M from #HorizonEurope, is setting the stage for innovation in Human-in-the-Loop AI to bridge the gap between intelligent systems and human decision-making across healthcare, justice, and policy. With 22 partners across 9 countries, TANGO is advancing AI that doesn’t replace humans—but empowers them. By integrating human oversight and collaboration, we’re building technology that prioritizes trust, autonomy, and collaboration. 🚀 The future of AI is human-driven, human-centered, and powered by meaningful human-AI collaboration! 🌐 Learn more about our vision and progress at: www.tango-horizon.eu What does Human-in-the-Loop mean to you? Let’s discuss! #HumanCentricAI #HumanInTheLoop #AIInnovation

TANGO ABC continuous! 💥🔤 K is for...reaching the KPIs! 🚀 The €7M #TANGO_Project is on course to meet its Key Performance Indicators, driving innovation in #HumanCentricAI under Horizon Europe. By focusing on: 🔍 Cognition-Aware Explainable AI for transparent decision-making. 🤝 Human-in-the-Loop Processes that foster collaboration. 💡 Cognitive Theories enhancing human-machine understanding. TANGO is delivering real-world impact in healthcare, justice, and public policy—fields where trust and transparency are essential. With 22 partners across 9 countries, TANGO is setting new standards for AI that works with us, for us. 🌍✨ #ArtificialIntelligence #Innovation #AIForGood #EthicalAI #HorizonEurope

📢 Recap: Algorithmic Authority & AI Influence in Decision Settings Workshop at HAI 2024 We are thrilled to share highlights from the successful half-day workshop on Algorithmic Authority & AI Influence in Decision Settings, held on November 24, 2024, organised by our project partners Swansea University as part of the 12th International Conference on Human-Agent Interaction (HAI 2024). ✨ Key Moments Keynote by Prof. Joel Fischer: A compelling presentation on agency, trust, and blame in Twitter discourses. His insights also explored fascinating new research on how humans discern AI advice from lawyers yet often favor AI input. Published work: https://lnkd.in/eYTZKZ2J Work in preparation on how humans may distinguish AI advice from lawyers but still favour AI advice: https://lnkd.in/enqSjdkA Speaker Highlights: Scholars from Italy, Switzerland, and the UK presented extended abstracts, sparking vibrant discussions on AI's growing role in decision-making contexts. 💡 Next Steps The productive and lively dialogue has inspired us to consider a follow-up event. Stay tuned for updates as we continue this important conversation! A huge thank you to Swansea University, the organizing team, and our co-organizers, Tango Partners, for making this workshop a success. Let’s keep pushing the boundaries of human-agent interaction! #HAI2024 #AlgorithmicAuthority #AIInfluence #HumanAgentInteraction #SwanseaUniversity

💥🔤 TANGO ABC continuous! J is for...#JointCollaboration. Together, as part of the Horizon Europe HORIZON-CL4-2022-HUMAN-02-01 cluster, with AI4REALNET Project, PEER AI, THEMIS 5.0, and the HUMAINE projects, we are working on advancing ethical, reliable #AISolutions and strengthening #HumanAICollaboration across diverse industries. Join our cluster as we work to establish AI standards that prioritize human values and needs. 👉 Read more in our Press Release: https://lnkd.in/e_teaVFZ #HumanAIcollaboration #AI #HorizonEurope #EUfunded #EthicalAI #Innovation

💥🔤 TANGO ABC continuous! I is for...Innovation At the heart of the €7M #HorizonEurope #Innovative TANGO Project lies a bold commitment, reshaping the future of AI to truly complement human decision-making. In high-stakes fields like healthcare, justice, and public policy, where trust and transparency are non-negotiable, TANGO is rewriting the rulebook for #HumanCentricAI. This is not just conceptual, our #Innovations are being tested in real-world scenarios, from empowering pregnant women, assisting surgical teams, to ensuring fairness in credit lending and public policy. Join us in shaping a future where AI and humans innovate together. #Innovation #InnovationForImpact #HumanCentricAI #HorizonProjects

📅 Recap from the Discovery Science 2024, October 14-16 in Pisa. Over the two days in October, an inspiring conference with nearly 150 researchers from Europe, Canada, and the U.S., gathering to discuss #ArtificialIntelligence, #MachineLearning, and #DataScience in scientific discovery, took place at our partners Università di Pisa. 🌟 Highlights: Roberto Navigli’s Keynote on gaps in today’s Large Language Models, covering reasoning, context retention, and biases. Carlos Castillo’s Talk on Human Factors and Algorithmic Fairness in decision-support systems. Francesca Toni’s Presentation on Explainable AI and Contestability, advocating for AI systems that support interactive explanations and revisable decisions. 🍽️ The event featured vibrant discussions, a social dinner, and a closing ceremony with Professor Sašo Džeroski, who announced Slovenia as the host of #DiscoveryScience2025. Thank you to all who made this conference a success! Excited for the future of #AI and #DataScience! #ArtificialIntelligence #Innovation #AIForGood #EthicalAI #HumanCognition #HumanAICollaborationModels #SynergisticSystems #HumanAISynergy #HorizonEurope

💥🔤 TANGO ABC marches on! H is for... #HumanCentricAI The #TANGO_Project, backed by €7M from #HorizonEurope, is leading #HumanCentricAI innovation to bridge AI with human decision-making in healthcare, justice, and policy. Our goals: 🔍 Explainable AI (XAI) 🤝 Human-in-the-Loop decision-making 💡 Cognitive theory for human-machine understanding With 22 partners across 9 countries, TANGO is pioneering AI that fosters trust, autonomy, and collaboration. 🚀 The future of AI is human-driven! 🌐 Learn more at: www.tango-horizon.eu #ArtificialIntelligence #Innovation #AIForGood #EthicalAI #HumanCognition #HumanAICollaborationModels #SynergisticSystems #HumanAISynergy #HorizonEurope