Skip to content
XAI group / LIACS
Explainable AI group · LIACS · Leiden University

We make complex systems understandable enough to trust.

Better decisions when structure is hidden, data is imperfect, and stakes are high — in engineering, infrastructure, sustainability and AI-driven design.

Fig. 1 — The intersection of Evolutionary Computation, XAI and Automated Algorithm Design.
01 — MISSION

Explainable, reproducible, and scientifically justified.

We develop and analyse methods that help people make better decisions, delivering models and tools that are not only powerful but trustworthy.

02 — VISION

One coherent research framework.

Connecting explainable AI, real-world-inspired benchmarking, class-specific optimisation and generative methods.

03 — CULTURE

Strong science and a healthy team are inseparable.

Ambitious work without sacrificing wellbeing or psychological safety. A group committed to explainability must also foster openness and trust.

Read our full mission and vision

The Explainable AI (XAI) group at LIACS studies intelligent systems in complex, partially known and fully opaque environments. Across predictive maintenance, black-box optimisation, multi-objective modelling, automated algorithm design and generative AI, we focus on one central goal: making complex systems understandable and reliable enough to trust.

Our mission

In the fast-paced world of advanced technology, where complex algorithms and models make decisions that affect our lives on a daily basis, it is imperative to understand and trust the reasoning behind these automated systems. We therefore develop and analyse methods that help people make better decisions when structure is hidden, data is imperfect, and stakes are high. Our work targets real-world domains such as engineering, infrastructure, sustainability, and AI-driven design.

We aim to deliver models and tools that are not only powerful, but also:

  • explainable,
  • reproducible,
  • and scientifically justified.

Our vision

Over the next years, we aim to build a coherent research framework connecting explainable AI, real-world-inspired benchmarking, class-specific optimisation and generative methods in order to advance the state-of-the-art in AI and solve real-world problems effectively and efficiently.

Alongside scientific progress, we are building a research culture where people can do ambitious work without sacrificing wellbeing or psychological safety. We believe strong science and a healthy team are inseparable.

Why this matters

Without this direction, AI risks becoming opaque and brittle: systems that look impressive but fail under real-world conditions. At the same time, research teams risk becoming high-pressure environments where people feel isolated or unsafe to speak openly.

We see both as unacceptable. A group committed to explainability must also foster openness, honesty, and trust in how it works together.

Group lead and collaborations

The group is led by Dr. Niki van Stein and is part of the Natural Computing Cluster (NACO) at LIACS, Leiden University. We work closely with academic and industry partners to ensure that our methods are both scientifically rigorous and practically relevant.

Open source

Our flagship open-source framework: LLaMEA evolves metaheuristics with large language models. Try it in one line.

$ pip install llamea
# LLaMEA · IEEE Transactions on Evolutionary Computation, 2024
$ git clone https://github.com/XAI-liacs/LLaMEA
People · 19

Who we are

Part of the Natural Computing Cluster (NACO) at LIACS, working with academic and industry partners.

  • Niki van Stein GROUP LEAD
  • Qi Huang RESEARCHER
  • Sofoklis Kitharidis RESEARCHER
  • Bernd Wagner RESEARCHER
  • Christiaan Lamers RESEARCHER
  • Kirill Antonov RESEARCHER
  • Alexander Zeiser RESEARCHER
  • Jiahuizi Luo RESEARCHER
  • Iryna Kovalchyk RESEARCHER
  • Ananta Shahane RESEARCHER
  • Tobias Preintner RESEARCHER
  • Farrukh Baratov RESEARCHER
  • Haoran Yin RESEARCHER
  • Annie Wong RESEARCHER
  • Suju Li RESEARCHER
  • Santiago Garcia Mayayo RESEARCHER
  • Aditya Bodas RESEARCHER
  • Roberta Tamponi RESEARCHER
  • Amir Sahrani RESEARCHER