Find below EMERGE’s news articles, press releases, digests of scientific publications, as well as other outreach materials.
Publication: Embedded deep reservoir computing for modelling complex industrial systems
17 October 2025
In this work, EMERGE partners from the University of Pisa propose the use of Deep Echo State Networks to model an industrial system.
Publication: Contact-Aware Safety in Soft Robots Using High-Order Control Barrier and Lyapunov Functions
15 October 2025
In this work, EMERGE partners from the Delft University of Technology introduce a comprehensive framework that enforces strict contact force limits across the entire soft-robot body during environmental interactions.
Publication: Direct Communication or Stigmergy? Selecting Communication Mechanisms for Robot Swarms via Automatic Modular Design
06 October 2025
In this paper, EMERGE partners from the University of Bristol show that automatic modular design (AutoMoDe) can also select between direct communication and stigmergy within a single design process.
Publication: Back to Bee-sics: Learning Information Sharing Strategies for Robot Swarms Through the Hive
06 October 2025
In this study, EMERGE partners from the University of Bristol use a learning-based approach to optimise information sharing in a hybrid robot swarm, where each robot maintains local autonomy, but information is shared via a central repository.
Publication: MAINLE: A Multi-Agent, Interactive, Natural Language Local Explainer of Classification Tasks
30 September 2025
In this work, EMERGE partners from the University of Pisa introduce a multi-agent architecture to provide interactive explanations for classification tasks based on a range of machine learning algorithms, so that end-users can obtain answers in natural language.
Publication: Sparse Autoencoders Find Partially Interpretable Features in Italian Small Language Models
26 September 2025
In this work, EMERGE partners from the University of Pisa provide an early evaluation on the feasibility of using Sparse Autoencoders to interpret models trained to be natively Italian.
Publication: MAIA: a Benchmark for Multimodal AI Assessment
26 September 2025
In this work, EMERGE partners from the University of Pisa introduce MAIA, a multimodal dataset developed as a core component of a competence-oriented benchmark designed for fine-grained investigation of the reasoning abilities of Visual Language Models (VLMs) on videos.
Publication: LLMs Struggle on Explicit Causality in Italian
26 September 2025
In this work, EMERGE partners from the University of Pisa present ExpliCITA, a translation of the English ExpliCa dataset
Publication: Benchmarking Nonlinear Readouts in Linear Reservoir Networks
25 September 2025
In this work, EMERGE partners from the University of Pisa address this gap by systematically benchmarking a spectrum of nonlinear readouts within linear RC frameworks.
