Human-Centred Machine Learning: Lighter, Clearer, Safer

Informazioni relative al progetto

ACHILLES

ID dell’accordo di sovvenzione: 101189689

DOI

10.3030/101189689

Data della firma CE 5 Novembre 2024

Data di avvio 1 Novembre 2024

Data di completamento 31 Ottobre 2028

Finanziato da

• Digital, Industry and Space

Costo totale

€ 9 616 256,25

Contributo UE

€ 8 226 276,25

Coordinato da

ASSOCIACAO FRAUNHOFER PORTUGAL RESEARCH
Portugal

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

Documents, reports (5)

Efficient AI Report v1

Preliminary report on methods for efficient training and inference of AI models, including finetuning, hyperparameter optimization, and recommendation on optimal infrastructure.Relevant tasks: T3.1, T3.2, T3.4

Legal and Ethical Mapping v1

Preliminary report on the mapping of the legal and ethical requirements of AI systems.Relevant tasks: T4.1

Dissemination and Communication Plan v1

Preliminary plan for Dissemination of Results and Communication activities.Relevant tasks: T8.1, T8.2

Data standardization and interoperability guidelines

Report on guidelines for data standardization and interoperability. Provides the framework for ensuring consistent and interoperable data usage across the project.Relevant tasks: T1.1

Privacy-preserving Learning Report v1

Preliminary report on privacy-preserving learning techniques including federated learning and secure computation strategies.Relevant tasks: T2.3

Pubblicazioni

Conference proceedings (15)

The Algorithm Is Not the Behavior: Learned Priors Override Look-Ahead in a Chess-Playing Neural Network

Autori: Elias Sandmann, Sebastian Lapuschkin, Wojciech Samek
Pubblicato in: 2025
Editore: NeurIps 2025 MechInterp Workshop

FUR: Fast and Unlimited Reads on Persistent Memory Transactions

Autori: João Barreto, Daniel Castro, Paolo Romano, Alexandro Baldassin
Pubblicato in: Proceedings of the 21st European Conference on Computer Systems, 2026
Editore: ACM
DOI: 10.1145/3767295.3769343

Towards On-the-Fly Snapshot Memory Compression for Low-Latency Elastic Inference Serving Systems

Autori: Radostin Stoyanov, Viktória Spišaková, Adrian Reber, Andrei Vagin, Rodrigo Bruno
Pubblicato in: Proceedings of the Sixth European Workshop on Machine Learning and Systems, 2026
Editore: ACM
DOI: 10.1145/3805621.3807612

See What I Mean? CUE: A Cognitive Model of Understanding Explanations

Autori: Tobias Labarta, Nhi Hoang, Katharina Weitz, Wojciech Samek, Sebastian Lapuschkin, Leander Weber
Pubblicato in: 2025
Editore: IJCAI 2025 Workshop on Explainable Artificial Intelligence (XAI), Montreal, Canada, August 2025

"""What's Happening in This Line?"" Eliciting Human Concepts and Mental Models in Time Series Interpretation"

Autori: Matilde Silva, Ricardo Melo, André V. Carreiro, Duarte Folgado
Pubblicato in: Proceedings of the Extended Abstracts of the 2026 CHI Conference on Human Factors in Computing Systems, 2026
Editore: ACM
DOI: 10.1145/3772363.3798546

Accelerating Transactional Execution via Processing-In-Memory

Autori: André Lopes, Daniel Castro, Paolo Romano
Pubblicato in: Proceedings of the 21st European Conference on Computer Systems, 2026
Editore: ACM
DOI: 10.1145/3767295.3803621

Navigating Ethical Dilemmas: The Value Landscape Behind Fieldwork Decisions

Autori: Inês Silva, Joana Couto da Silva, Francisco Nunes, Ricardo Melo
Pubblicato in: Proceedings of the 2026 CHI Conference on Human Factors in Computing Systems, 2026
Editore: ACM
DOI: 10.1145/3772318.3790843

Structural Compactness as a Complementary Criterion for Explanation Quality

Autori: Mohammad Mahdi Mesgari, Jackie Ma, Wojciech Samek, Sebastian Lapuschkin, Leander Weber
Pubblicato in: 2025
Editore: XAI World Conference 2026

Manipulating Feature Visualizations with Gradient Slingshots

Autori: Dilyara Bareeva Marina M.-C. Höhne, Alexander Warnecke,Lukas Pirch, Klaus-Robert Müller,Konrad Rieck,Sebastian Lapuschkin,Kirill Bykov
Pubblicato in: 2025
Editore: NeurIps 2025

STRATEGIC DISHONESTY CAN UNDERMINE AI SAFETYEVALUATIONS OF FRONTIER LLMS

Autori: Alexander Panfilov, Evgenii Kortukov, Kristina Nikoli, Matthias Bethge, Sebastian Lapuschkin, Wojciech Samek, Ameya Prabhu, Maksym Andriushchenko, Jonas Geiping
Pubblicato in: 2026
Editore: ICLR2026

ATLAS-ALIGNMENT: MAKING INTERPRETABILITY TRANSFERABLE ACROSS LANGUAGE MODELS

Autori: Bruno Puri, Jim Berend, Sebastian Lapuschkin, Wojciech Samek
Editore: ICLR2026

Engine-Agnostic Model Hot-Swapping for Cost-Effective LLM Inference

Autori: Radostin Stoyanov, Viktória Spišaková, Adrian Reber, Wesley Armour, Marcin Copik, Rodrigo Bruno
Pubblicato in: Proceedings of the SC '25 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis, 2026
Editore: ACM
DOI: 10.1145/3731599.3767354

Flexico: Sustainable Machine Translation via Self-Adaptation

Autori: Maria Casimiro, Paolo Romano, José G. C. De Souza, Amin M. Khan, David Garlan
Pubblicato in: 2025 IEEE/ACM 20th Symposium on Software Engineering for Adaptive and Self-Managing Systems (SEAMS), 2025
Editore: IEEE
DOI: 10.1109/SEAMS66627.2025.00024

Poster: Secure Lifecycle Management of Confidential Virtual Machines in Public Clouds

Autori: João Sereno, Daniel Castro, Nuno Santos, Luis Rodrigues
Pubblicato in: 2025 23rd International Symposium on Network Computing and Applications (NCA), 2025
Editore: IEEE
DOI: 10.1109/NCA67271.2025.00057

Better Memory Tiering, Right from the First Placement

Autori: João Póvoas, João Barreto, Bartosz Chomi?ski, André Gonçalves, Fedar Karabeinikau, Maciej Maciejewski, Jakub Schmiegel, Kostiantyn Storozhuk
Pubblicato in: Proceedings of the 16th ACM/SPEC International Conference on Performance Engineering, 2025
Editore: ACM
DOI: 10.1145/3676151.3719378