Institution profile

Université Côte d’Azur

Academic institutioneurope · fr
Official website
Research library282linked papers
Opportunities0open roles
Selected work

Representative Papers

Minimum Star Partitions of Simple Polygons in Polynomial Time

Nov 17, 2023Symposium on the Theory of Computing

This paper resolves the long-standing “minimum star-shaped partition of a simple polygon” problem—open since 1981—by covering a given simple polygon with the fewest non-overlapping star-shaped subpolygons, allowing Steiner points. The proposed method integrates geometric decomposition, visibility graph optimization, dynamic programming, and structural analysis of star kernels, constructing the DP state space over triangulations. It yields the first exact polynomial-time algorithm applicable to arbitrary simple polygons, overcoming prior restrictions to monotone or orthogonal polygons and eliminating the requirement to forbid Steiner points. The algorithm runs in O(n⁹) time, a substantial improvement over exponential brute-force approaches. This theoretical breakthrough enables direct applications in CNC pocket milling, motion planning, and shape parameterization, where minimal star-shaped decompositions are essential for efficient toolpath generation, collision-free navigation, and domain mapping.

4 citations1 influentialRead paper

On the Dynamics of Bounded-Degree Automata Networks

Nov 14, 2025

This paper investigates the dynamical realizability problem for finite automata networks over bounded-degree communication graphs, focusing on constructing and characterizing canonical dynamics such as single-cycle behavior and Gray-code mappings. Methodologically, it integrates graph theory, finite dynamical systems theory, and isomorphism analysis, leveraging explicit cycle-structure constructions and complexity-theoretic reductions. The work establishes fundamental impossibility results: several key dynamical parameters—including the number of fixed points and network rank—are provably unrealizable under bounded-degree constraints. Moreover, it constructs, for the first time, a minimal realization exhibiting exactly one fixed point and a single cycle covering all remaining configurations. Collectively, these results precisely characterize the dynamical capability frontier of bounded-degree networks, establish tight computational complexity bounds (both upper and lower) for associated realizability decision problems, and provide foundational theoretical support for modeling distributed systems under topological constraints.

1 citationsRead paper

Kastor: Fine-Tuned Small Language Models for Shape-Based Active Relation Extraction

Nov 05, 2025Extended Semantic Web Conference

This work addresses the low accuracy and poor generalization of relation extraction in domain-specific knowledge bases. We propose a fine-tuning method for small language models (SLMs) guided jointly by RDF schemas and SHACL shapes. Methodologically, we extend conventional single-attribute SHACL validation to multi-attribute compositional constraint modeling and integrate an iterative active learning mechanism, enabling fine-grained relation identification and novel fact discovery under limited annotated text and RDF data. Our key contributions are: (1) the first deep integration of SHACL shape semantics into SLM fine-tuning, endowing the model with schema-aware reasoning capabilities; and (2) enhanced noise robustness and generalization via attribute-coordinated evaluation. Experiments on knowledge base completion demonstrate that our approach significantly outperforms baseline models, achieving a 12.7% absolute improvement in relation extraction F1-score and effectively discovering domain-specific novel relations.

1 citationsRead paper

On-the-fly Reconstruction for Large-Scale Novel View Synthesis from Unposed Images

Jun 05, 2025

Existing methods for novel view synthesis from large-scale unstructured image sequences suffer from high computational latency, memory bottlenecks, and SLAM failure under wide-baseline or large-scale scenarios—particularly during camera pose estimation and 3D Gaussian optimization. This paper introduces the first real-time, online dynamic Gaussian radiance field framework that enables simultaneous capture and reconstruction. Our method jointly optimizes camera poses and the radiance field via learning-based fast initial pose estimation and a GPU-efficient miniature bundle adjustment. We propose a novel dynamic incremental Gaussian primitive generation scheme coupled with anchor-point clustering and offloading to alleviate memory and computational constraints. Furthermore, direct Gaussian sampling and progressive anchor storage enhance rendering efficiency. Evaluated across diverse datasets, our approach achieves “reconstruction-on-the-fly,” significantly outperforming offline methods in processing speed while matching state-of-the-art rendering quality—fully supporting dense, wide-baseline, and ultra-large-scale scenes.

1 citationsRead paper

Computing Non-Obtuse Triangulations with Few Steiner Points

May 29, 2025

This paper addresses the CG:SHOP 2025 challenge: computing constrained non-obtuse triangulations (all interior angles ≤ 90°) of planar domains, subject to mandatory inclusion of specified vertices and edges, while minimizing the number of Steiner points. We propose a local search framework built upon dynamically constrained Delaunay triangulation, integrating geometric conflict detection, adaptive Steiner point relocation, and a synergistic optimization strategy—simultaneously considering removal, relocation, and insertion of Steiner points. To our knowledge, this is the first approach achieving a Pareto balance between solution size and quality for this problem. Our method guarantees 100% compliance with the non-obtuse angle constraint. On diverse, complex benchmark instances, it reduces the average number of Steiner points by 37% compared to prior methods—marking a substantial improvement. The approach secured first place in the CG:SHOP 2025 competition.

1 citationsRead paper
Recent publications

Latest Papers