coconut banner
coconut banner coconut banner coconut banner

Welcome to Coconut

Coconut is a research group in computer science at the LIRMM laboratory in Montpellier, France. We are interested in artificial intelligence, with a primary focus on constraint programming.

Research topics

Foundations of constraint programming

- Propagation, search strategies, spatio-temporal constraints, interval constraint programming, complexity

Constraints and learning

- Constraint acquisition, learning for search

Applications

- Digital agriculture, robotics, automotive industry

Members

Permanent Members

Students

  • Yoan THOMAS, PhD Student

Other (Postdocs...)

  • Zouhayra AYADI, ATER
You can also see our former members here:

Talks and Seminars

Talks and Seminars
Constraint programming for space missions: data transfers and tasks scheduling
2026-04-10 15:30
E323
Julien Rouzot
In the context of space missions, many decisions related to the ground segment are made manually or semi-manually by teams of experts in collaboration with the scientific team. This process is time-consuming, produces suboptimal solutions, and may reduce the scientific return of the mission. We present two combinatorial optimization problems from real-world cases: one related to data transfers for the Rosetta mission, and the other to the periodic scheduling of tasks for the NIMPH nanosatellite mission. We show that both problems are NP-hard and propose constraint programming models to solve them. The first model incorporates new global constraints and a dedicated search strategy, while the second is integrated into a software that enables the NIMPH team to generate and visualize diverse solutions that can be directly used in the nanosatellite’s onboard software. Our experimental results show that our CP-based approaches improve the quality of the solutions compared to previous approaches and are able to provide optimality certificates in some non-trivial cases.
BFS-Based Canonical Codes for Generating Graphs with CP
2026-03-27 15:30
E323
Xiao Peng
In combinatorics and graph theory, graph enumeration refers to systematically searching for all non-isomorphic graphs that satisfy some predefined properties. To avoid generating redundant graphs, canonical codes are used to identify the isomorphism class of a graph. A widely used canonical code is based on the adjacency matrix as its lexicographically minimal form is unique to represent a graph. In this work, we adopt a constraint programming approach to graph enumeration. We investigate an alternative encoding by listing all the edges following a certain traversal order. In particular, we show that BFS-based traversals can be naturally encoded by means of constraints in the construction of canonical codes. We also introduce a global constraint that breaks all symmetries up to isomorphism. We show that this new encoding can be efficiently applied to generate claw-free cubic graphs and extremal graphs with required girth.
Programmation par contraintes pour la génération de texte sous contraintes
2025-12-19 15:30
E323
Alexandre Bonlarron
Produire un texte naturel tout en respectant des contraintes globales et non locales (longueur, vocabulaire imposé, ponctuation, syllabes, mise en page, etc.) reste un défi : les modèles de langue (LM/LLM) génèrent des textes plausibles, mais ne garantissent pas la conformité à des spécifications strictes. Cette présentation introduit deux approches qui combinent un « Système 1 » prédictif (LM/LLM) et un « Système 2 » assertif (programmation par contraintes) afin de générer des textes lisibles avec garanties. Dans Contraindre-puis-Prédire (Constraint First), l’ensemble des textes admissibles est décrit par une modélisation par contraintes (CSP/MDD), puis les solutions sont classées à l’aide d’un modèle de langue. Dans Prédire-et-Contraindre (GenCP), un LLM guide la recherche (choix de domaines), tandis que la propagation de contraintes et le retour arrière assurent la validité ; un modèle de langue masqué fournit une prévisualisation bidirectionnelle pour mieux gérer les contraintes distantes. Les méthodes seront illustrées sur la génération de tests de lecture sensibles à la mise en page et sur le benchmark COLLIE, où elles montrent une fiabilité supérieure à celle d’un décodage classique (glouton) suivi de vérifications a posteriori.
De la programmation par contraintes à l'exploration sous-marine
2025-12-12 14:30
E323
Simon Rohou
La programmation par contraintes et l’analyse par intervalles constituent un cadre unifié pour résoudre de manière élégante des problèmes difficiles de robotique mobile. En particulier, le problème de localisation et de cartographie simultanées (SLAM) reste un sujet d’étude et présente de nombreux challenges, notamment en contexte sous-marin. Les capteurs des robots génèrent des informations hétérogènes qui sont autant de contraintes discrètes ou continues provenant des observations de l’environnement, ou de contraintes différentielles issues de la dynamique. Elles sont souvent associées à de fortes incertitudes temporelles et à des mesures ambiguës. En formulant ces éléments sous forme de contraintes, il devient possible d’enfermer rigoureusement toutes les solutions compatibles, telles que l’ensemble des trajectoires possibles d’un robot, et de traiter simplement des situations qui mettent souvent en échec les méthodes conventionnelles de localisation. Nous illustrerons cette approche sur des données réelles, avec de la « localisation référencée terrain » sur la base de cartes bathymétriques, puis avec du SLAM sur des textures du fond marin.

Projects

AudioMobility 20302024-2026

Le projet AM 2030 permettra aux constructeurs automobiles de disposer de leur propre application audio embarquée, quel que soit le système d’exploitation. Ils pourront déployer une expérience globale audio et offrir les meilleurs contenus et services proactifs aux automobilistes. Il se positionne comme un véritable compagnon de route qui aidera le consommateur à adopter des comportements éco-responsables : auto-diagnostique du véhicule et rapport de maintenance, conseil quant à la conduite et l’usage des équipements à bord. Partenaires : ETX Studio, CONTINENTAL Automotive, Université Fédérale Toulouse, École Polytechnique

Contact: Christian Bessiere

TAILOR (Developing the scientific foundations for Trustworthy AI through the integration of Learning, Optimisation and Reasoning)2020-2024

The purpose of the EU Project TAILOR is to build the capacity to provide the scientific foundations for Trustworthy AI in Europe by developing a network of research excellence centres leveraging and combining learning, optimisation, and reasoning. These systems are meant to provide descriptive, predictive, and prescriptive systems integrating data-driven and knowledge-based approaches. Artificial Intelligence (AI) has grown in the last ten years at an unprecedented pace. It has been applied to many industrial and service sectors, becoming ubiquitous in our everyday life. More and more often, AI systems are used to suggest decisions to human experts, to propose actions, and to provide predictions. Because these systems might influence our life and have a significant impact on the way we decide, they need to be trustworthy. How can a radiologist trust an AI system analysing medical images? How can a financial broker trust an AI system providing stock price predictions? How can a passenger trust a self-driving car? These are fundamental questions that require deep analysis and fundamental research activity as well as a new generation of AI talents who are skilled in the scientific foundations of Trustworthy AI, who know how to assess, and how to design, trustworthy AI systems. Some of the current issues related to lack of trust in AI systems are a direct consequence of the massive use of black-box methods relying only on data. We need to define the foundations of a new generation of AI systems not only relying on data-driven approaches, but also on the whole set of AI techniques, including symbolic AI methods, optimization, reasoning, and planning.

Contact: Christian Bessiere

Hybrid AI Systems Grounded on Spatio-Temporal Reasoning2023-2028

In the proposed research project, I, together with collaborators from the Coconut team and the rest of my research network, will work on the interplay between knowledge representation and machine learning, especially in the context of attaining spatiotemporally-informed AI frameworks. Of particular importance for this integration of learning and reasoning will be the area of Qualitative Spatial and Temporal Reasoning, or QSTR for short, which is a major field of study in AI that deals with the fundamental cognitive concepts of space and time in a symbolic, human-like manner. Simply put, QSTR abstracts from numerical quantities of space and time by using qualitative descriptions instead (e.g., precedes, contains, is left of ), with applications in a plethora of areas and domains such as smart environments, intelligent vehicles, and unmanned aircraft systems. The project is tied to my Junior Professorship Chair (CPJ) and is funded by ANR and the the I-SITE program of excellence of the University of Montpellier.

Contact: Michael Sioutis

Artificial Intelligence for Health and Environment – AXIAUM2021-2024

The AXIAUM project focuses on raising innovative interdisciplinary research topics on health and environment related to artificial intelligence (AI) with the scope of developing both new methodologies and finalized research. The ambition of the programme proposed by the consortium University of Montpellier/IMT Mines Alès is to strongly link new research on AI to the existing research and the AI strategy of the consortium, reinforced by the current structuration of data science within the ISDM Montpellier Institute of Data Science.

Contact: Clément Carbonnel