James Hinns

Postdoctoral researcher in human-centred interpretability in the Augment group at KU Leuven.

website_ny_snow.jpg

I recently completed my PhD in Explainable AI with the Applied Data Mining group at the University of Antwerp.

My research focuses on human-centred interpretability, explanation evaluation, and machine learning systems that are more understandable, trustworthy, and aligned with human needs.

news

September 2026 Publicly defended my PhD thesis, Human-Centred Approaches for Generating and Evaluating Explainable AI Across Modalities, at the University of Antwerp, completing my PhD.
July 2026 Our paper, Aggregating Local Saliency Maps for Semi-Global Explainable Image Classification, was accepted for presentation at the XKDD and Beyond Workshop at ECML PKDD 2026.
June 2026 Our article, On the Definition and Detection of Cherry-Picking in Counterfactual Explanations, was accepted through the ECML PKDD 2026 Journal Track. It will appear in Data Mining and Knowledge Discovery and be presented at ECML PKDD in September 2026.
February 2026 Our article, Cash or Comfort? How LLMs Value Your Inconvenience, was accepted for publication in Communications of the ACM.

latest posts

selected publications

  1. On the Definition and Detection of Cherry-Picking in Counterfactual Explanations
    James Hinns, Sofie Goethals, Stephan Veeken, and 2 more authors
    Data Mining and Knowledge Discovery, 2026
  2. Exposing Shortcuts in Image Classification by Aggregating Counterfactuals
    James Hinns and David Martens
    In Computational Intelligence, 2026
  3. Aggregating Local Saliency Maps for Semi-Global Explainable Image Classification
    James Hinns and David Martens
    In XKDD and Beyond Workshop at ECML PKDD 2026, 2026
    Accepted for presentation
  4. DSS
    Tell Me a Story! Narrative-Driven XAI with Large Language Models
    David Martens, James Hinns, Camille Dams, and 2 more authors
    Decision Support Systems, 2025