This thesis aims to conduct an experimental study of an unstable excited state of the helium-4 nucleus, in order to gain a better understanding of its structure and decay modes. Using a specialised detector and a neutron beam at GANIL, the PhD student will measure the various decay channels and their respective probabilities. The results will enable the testing of theoretical models describing the behaviour of nuclei in an unstable state.

This thesis aims to develop a new laser spectroscopy technique to study the atomic and nuclear properties of superheavy elements. It focuses primarily on the design of an experimental instrument and a cryogenic cell capable of detecting and analysing extremely rare and unstable atoms, before applying this to the study of lawrencium and other superheavy elements

The next generation of neutrino experiments aims to achieve unprecedented precision in neutrino interaction cross-section measurements, a prerequisite for reducing systematic uncertainties in future oscillation experiments such as DUNE and Hyper-Kamiokande. The proposed nuSCOPE facility at CERN introduces an innovative approach based on monitored and individually tagged neutrino beams, where precise reconstruction of charged particles produced in meson decays provides information on the neutrino flux, flavour and energy. Realising this concept requires detector systems combining excellent timing and spatial resolution, high-rate capability and large-area coverage.

This PhD project proposes to investigate precision timing and particle association using complementary gaseous and silicon detector technologies. The main experimental activity will focus on developing a new generation of large-area PICOSEC-Micromegas detectors, exploiting Cherenkov-light conversion and fast gaseous amplification to achieve timing resolutions of a few tens of picoseconds. Particular attention will be devoted to understanding the physical mechanisms limiting timing performance and efficiency, optimising robust photocathodes, detector geometry and readout electronics, and preserving uniform performance when scaling towards larger surfaces. The integration of PICOSEC timing layers with thin Micromegas tracking detectors will also be investigated. In parallel, the project will contribute to the characterisation of MiniCACTUS monolithic silicon sensors, combining fine spatial granularity, precise timing and high-rate capability for the central regions of the spectrometer.

A central scientific objective will be to establish the relationship between detector performance and neutrino-tagging capabilities. Through simulations and reconstruction studies, the doctoral researcher will investigate how timing resolution, spatial granularity, detection efficiency and particle multiplicity influence track association, accidental coincidences and neutrino reconstruction. These studies will provide quantitative feedback for optimising the complementary detector technologies.

The experimental programme will exploit dedicated laser, cosmic-muon and radioactive-source facilities at CEA, enabling detector development during CERN's accelerator long shutdown. Supported by the recently funded ANR EXTASE project, providing resources for prototypes, equipment and scientific exchanges, the thesis will combine innovative detector development with physics-driven optimisation, contributing to nuSCOPE and future high-intensity particle-physics instrumentation.

This PhD topic is part of the NUCLEUS experiment, installed at the heart of the Chooz nuclear power plant in France, with an ambitious goal: detecting and measuring coherent elastic neutrino-nucleus scattering (CEvNS) at a nuclear reactor. At MeV energies, CEvNS is the dominant interaction mode of neutrinos with matter, but the very low-energy nuclear recoils it produces long made it unobservable: it was first measured in 2017, forty years after its prediction, at the Oak Ridge accelerator in the USA. Measuring it at a reactor is a twofold challenge: the nuclear recoils, with energies of the order of 100 eV, escape conventional detectors, and background conditions there are particularly unfavorable.
To meet these challenges, the NUCLEUS collaboration has developed cryogenic calorimeters: calcium tungstate crystals with a total mass of about 10 grams, operated at ~10 mK and read out by transition-edge sensors (TES), achieving detection thresholds of the order of 10 eV. These detectors are shielded from external backgrounds by an instrumented cryogenic shield, complemented by a radiological shield and a muon veto outside the cryostat, all designed and developed by our team. These ultra-low thresholds have however revealed a new limitation: a poorly understood instrumental background, dubbed the "low-energy excess" (LEE), whose rate rises exponentially below the keV and inders the expected CEvNS signal. Observed by all low-threshold cryogenic detection experiments, the LEE is today the main factor limiting the cryogenic detection of reactor antineutrinos, and understanding it is the guiding thread of this thesis.
Following a successful installation in 2026, NUCLEUS has entered its first data-taking phase, called the "technical run", using a simplified cryogenic detection module. It will be followed in 2027 by a "neutrino run", exploiting an updated version of the module integrating an instrumented crystal-holding system, called the inner veto, developed at the Technical University of Munich. During the thesis, the PhD student will take part in these data-taking campaigns and their analysis, mobilizing all the sub-detectors of the instrument. The first task will be to characterize particle backgrounds in the keV region in order to refine the Geant4 model developed by our team, an essential prerequisite to properly isolate the temporal and energy characteristics of the LEE in low-energy reconstructed events. A second part of the topic will then be dedicated to the characterization of the LEE and to the identification of its origin among the current hypotheses (mechanical stress induced by the crystal holder, relaxation of crystal defects), through a detailed comparison of the two detector configurations. This comparison will be decisive to test the mechanical stress hypothesis and, where appropriate, to reject the associated events. Combining these two campaigns will make it possible to better identify the experimental parameters that modulate the LEE event rate, and to guide the design of a new detector minimizing this background. The goal is to pave the way for a CEvNS measurement at a reactor with world-class sensitivity, with implications for neutrino physics and the monitoring of nuclear facilities.

The precise measurement of the Higgs boson's properties is a major challenge in particle physics and a fundamental test of the Standard Model. This thesis proposes to improve the measurement of its mass and the constraints on its natural width in the diphoton decay channel (H ? ??), by exploiting data from Run 3 of the CMS experiment at the LHC.
The project relies on developing new artificial intelligence-based calibration methods for the CMS electromagnetic calorimeter. Building on the IJazZ2.0 method developed at CEA/Irfu, the goal is to replace a calibration relying on a few hundred parameters with a continuous, multidimensional correction. Modern generative models will enable the rapid simulation of Z ? ee decays under kinematic constraints and improve both the energy scale and resolution of electrons and photons.
This thesis will thus combine artificial intelligence, instrumental physics, and precision measurements to deepen our understanding of the fundamental properties of the Higgs boson.

Since the discovery of the Higgs boson in 2012, the experimental program at the Large Hadron
Collider (LHC) has entered an era of precision measurements, enabling increasingly stringent tests of the
Standard Model (SM). While all measured Higgs boson properties remain consistent with SM predictions, a central parameter of the scalar sector is essentially unconstrained: the Higgs boson self-coupling, ?H.
This parameter governs the shape of the Higgs potential and is intimately connected to the (meta)stability
of the electroweak vacuum and the long-term evolution of the Universe. Its measurement therefore
represents one of the major outstanding challenges of the present and future LHC physics programme.
By combining the exploitation of the full Run-3 dataset with the development of next-generation trigger
algorithms for the HL-LHC, this PhD Thesis Project establishes a direct connection between precision Higgs
measurements and detector innovation. Beyond improving the sensitivity of the current HH ?bbtt analysis,
the project aims to identify and overcome future key experimental limitations, developing new reconstruction
and selection strategies that will contribute to shaping the future CMS physics programme.

BINGO is an innovative neutrino physics project designed to lay the groundwork for a large-scale bolometric experiment dedicated to the search for neutrinoless double beta decay. The goal is to achieve an extremely low background index—on the order of 10^-5 counts/(keV·kg·yr)—while delivering excellent energy resolution in the region of interest. These performance levels will enable the exploration of lepton number violation with unprecedented sensitivity.

The project relies on scintillating bolometers, which are particularly effective at rejecting the dominant background caused by surface alpha particles. It focuses on two highly promising isotopes, 100Mo and 130Te, whose complementary properties make them both strong candidates for future large-scale investigations.

BINGO introduces three major innovations to the well-established heat-light hybrid bolometer technology. First, the sensitivity of the light detectors will be enhanced by an order of magnitude through the use of Neganov-Luke amplification. Second, a novel detector assembly design will reduce surface radioactivity contributions by at least an order of magnitude. Third, and for the first time in a macrobolometer array, an internal active shield made of ultrapure BGO scintillators with bolometric light readout will be implemented to suppress external gamma background.

As part of this thesis work, the student will take part in the assembly and installation of the active cryogenic veto for the MINI-BINGO demonstrator within the cryostat recently installed at the Modane Underground Laboratory. He/she will be involved in data acquisition and analysis, and will contribute to evaluating the final background rejection enabled by the performance of the detector's final configuration.

Top-quark physics at the LHC offers two complementary test beds for quantum computing. The first is quantum information in top-quark pairs: since the top quark decays before hadronisation, a tt¯ pair behaves as a two-qubit system whose spin density matrix can be measured. This is a quantum-native problem with a known ground truth. The thesis will consist of establishing a full tomography pipeline with the ATLAS Run 3 data at 13.6 TeV, using quantum circuits, optimisation methods and a quantum parton shower to validate and study the measurements. The second aspect is linked to four-top production, a rare signal with challenging backgrounds, small training samples and a combinatorial many-jet final state. The project will test quantum optimisation (QUBO-based jet-to-parton assignment) and quantum classifiers for signal discrimination against the best classical methods, to find out where quantum approaches bring a real gain in accuracy, data efficiency or scaling.

In the Standard Model (SM), the Higgs field breaks the electroweak symmetry, thereby giving mass to the W and Z bosons. The self-coupling of the Higgs boson remains unmeasured, yet it is crucial for understanding the shape of the Higgs potential and the stability of the universe’s vacuum. Studying Higgs pair production (di-Higgs) is the only direct way to access this parameter, providing key insights into the electroweak phase transition after the Big Bang. Di-Higgs production is extremely rare (cross-section ~40 fb for proton-proton collisions at a centre-of-mass energy of 13.6 TeV), and among its possible final states, the multilepton channel (channel with light leptons) is promising due to its distinctive kinematics, though complex due to diverse topologies and backgrounds. Recent advances in artificial intelligence, particularly transformer-based architectures respecting physical symmetries, have recently significantly improved event reconstruction in complex Higgs channels such as HH?4b or HH?bbtt. Applying these techniques to the multilepton channel offers strong potential to enhance sensitivity. This PhD project will focus on searching for di-Higgs production in the multilepton final state with the full ATLAS Run 3 dataset at 13.6 TeV, leveraging the group’s ongoing ttH multilepton work to develop advanced AI-based reconstruction and analysis methods. The projet aims to approach SM sensitivity for the Higgs self-coupling.