José A. Galindo, researcher at Diverso Lab and the Department of Computer Languages and Systems of the University of Seville, has received a Most Influential Paper (MIP) award at VARIABILITY 2026, held in Limassol, Cyprus, from 29 September to 2 October.
The award recognises “Using Machine Learning to Infer Constraints for Product Lines” (Paul Temple, José A. Galindo, Mathieu Acher and Jean-Marc Jézéquel), presented at SPLC 2016 in Beijing.
https://dl.acm.org/doi/abs/10.1145/2934466.2934472
What sets an MIP apart is its timing. It is not awarded when a paper is published, but ten years later, when a committee can look back and judge which contributions actually shaped the field.
The paper addressed a long-standing problem in variability-intensive systems: finding a set of constraints that is both precise, allowing every correct configuration, and complete, never allowing a wrong one.
Instead of relying only on hand-written constraints, the authors proposed sampling configurations, checking the resulting products against an oracle, and using machine learning to infer the constraints from the results.
They validated the approach on a configurable video generator, with a computer-vision algorithm acting as the oracle. A decade later, combining machine learning and variability modelling is one of the most active research directions in the community, and this paper was among the first to explore it.
The work was carried out during José’s postdoctoral stay in France, with colleagues from Inria and the University of Rennes. It is a good example of how international research stays pay off in the long run.
“When we wrote this paper, machine learning for configurable systems was just getting started. Seeing how far the community has taken the idea is the best part of this award, and I share it with Paul, Mathieu and Jean-Marc,” says José.
The award also comes at a meaningful moment for the community: VARIABILITY 2026 is the first edition of the conference born from the merger of SPLC, VaMoS and ICSR. It is also the second MIP for our group, after “Automated Reasoning on Feature Models” by David Benavides, Pablo Trinidad and Antonio Ruiz Cortés (CAiSE 2005) was recognised at SPLC 2017 in Seville.
Congratulations to José, Paul, Mathieu and Jean-Marc!