Search

The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought. These advanced reasoning capabilities provide new avenues for improving the safety and robustness of our models. In particular, our models can reason about our safety policies in context when responding to potentially unsafe prompts, through deliberative alignment. This leads to state-of-the-art performance on certain benchmarks for risks such as generating illicit advice, choosing stereotyped responses, and succumbing to known jailbreaks. Training models to incorporate a chain of thought before answering has the potential to unlock substantial benefits, while also increasing potential risks that stem from heightened intelligence. Our results underscore the need for building robust alignment methods, extensively stress-testing their efficacy, and maintaining meticulous risk management protocols. This report outlines the safety work carried out for the OpenAI o1 and OpenAI o1-mini models, including safety evaluations, external red teaming, and Preparedness Framework evaluations.

In the quantum optimisation paradigm, variational quantum algorithms face challenges with hardware-specific and instance-dependent parameter tuning, which can lead to computational inefficiencies. However, the promising potential of parameter transferability across problem instances with similar local structures has been demonstrated in the context of the Quantum Approximate Optimisation Algorithm. In this paper, we build on these advancements by extending the concept to annealing-based protocols, employing Bayesian optimisation to design robust quasi-adiabatic schedules. Our study reveals that, for Maximum Independent Set problems on graph families with shared geometries, optimal parameters naturally concentrate, enabling efficient transfer from smaller to larger instances. Experimental results on the Orion Alpha platform validate the effectiveness of our approach, scaling to problems with up to $100$ qubits. We apply this method to address a smart-charging optimisation problem on a real dataset. These findings highlight a scalable, resource-efficient path for hybrid optimisation strategies applicable in real-world scenarios., Comment: 22 pages, 12 figures

Inertial waves in convective regions of stars exhibit topological properties linked to a Chern number of 1. The first of these is a unique, unidirectional, prograde oscillation mode within the cavity, which propagates at arbitrarily low frequencies for moderate azimuthal wavenumbers. The second one are phase singularities around which the phase winds in Fourier space, with winding numbers of $\pm 1$ depending on the hemisphere. Phase winding is a collective effect over waves propagating in all directions that is strongly robust to noise. This suggests a topology-based method for wave detection in noisy observational data., Comment: 8 pages, 9 figures. Accepted for publication in Physical Review Research

This is my (reviewed) PhD manuscript. It contains 6 Chapters, which contains mostly already published work, except for Chapter 5 which is new. Chapter 1 introduce basic notions on fractal geometry: the Fourier dimension, the thermodynamical formalism and additive combinatorics. Chapter 2 is a generalized version of arXiv:2211.08088. Chapter 3 is a slightly upgraded version of arXiv:2112.00701. Chapter 4 is a generalized version of arXiv:2301.10623. Chapter 5 and Chapter 4 together contains the first proof of the positivity of the Fourier dimension for basic sets of nonlinear, area-preserving, smooth Axiom A diffeomorphisms on surfaces. This is possible by adapting some ideas that can be found in Tsujii-Zhang's work arXiv:2006.04293. Some future work still needs to be done to prove that the nonlinearity conditions are generic in this setting: this should become a proper research article in the future. Chapter 6 is a slighty upgraded version of arXiv:2307.10755., Comment: 174 pages, PhD thesis

Electricity forecasting has been a recurring research topic, as it is key to finding the right balance between production and consumption. While most papers are focused on the national or regional scale, few are interested in the household level. Desegregated forecast is a common topic in Machine Learning (ML) literature but lacks explainability that household energy forecasts require. This paper specifically targets the challenges of forecasting electricity use at the household level. This paper confronts common Machine Learning algorithms to electricity household forecasts, weighing the pros and cons, including accuracy and explainability with well-known key metrics. Furthermore, we also confront them in this paper with the business challenges specific to this sector such as explainability or outliers resistance. We introduce a custom decision tree, aiming at providing a fair estimate of the energy consumption, while being explainable and consistent with human intuition. We show that this novel method allows greater explainability without sacrificing much accuracy. The custom tree methodology can be used in various business use cases but is subject to limitations, such as a lack of resilience with outliers.

Spin-based semiconductor qubits hold promise for scalable quantum computing, yet they require reliable autonomous calibration procedures. This study presents an experimental demonstration of online single-dot charge autotuning using a convolutional neural network integrated into a closed-loop calibration system. The autotuning algorithm explores the gates' voltage space to localize charge transition lines, thereby isolating the one-electron regime without human intervention. In 20 experimental runs on a device cooled to 25mK, the method achieved a success rate of 95% in locating the target electron regime, highlighting the robustness of this method against noise and distribution shifts from the offline training set. Each tuning run lasted an average of 2 hours and 9 minutes, primarily due to the limited speed of the current measurement. This work validates the feasibility of machine learning-driven real-time charge autotuning for quantum dot devices, advancing the development toward the control of large qubit arrays., Comment: 6 pages (main) + 5 pages (supplementary)

We introduce in this paper a definition of (non necessarily positive) opetopes where faces are organised in a poset. Then we show that this description is equivalent to that given in terms of constellations by Kock, Joyal, Batanin and Mascari.

The random motion of molecules in living cells has consistently been reported to deviate from standard Brownian motion, a behavior coined as ``anomalous diffusion''. Fluorescence Correlation Spectroscopy (FCS) is a powerful method to quantify molecular motions in living cells but its application is limited to a subset of random motions and to long acquisition times. Here, we propose a new analysis approach that frees FCS of these limitations by using machine learning to infer the underlying model of motion and estimate the motion parameters. Using simulated FCS recordings, we show that this approach enlarges the range of anomalous motions available in FCS. We further validate our approach via experimental FCS recordings of calibrated fluorescent beads in increasing concentrations of glycerol in water. Taken together, our approach significantly augments the analysis power of FCS to capacities that are similar to the best-in-class state-of-the-art algorithms for single-particle-tracking experiments.

Experimentally monitoring the kinematics of branching network growth is a tricky task, given the complexity of the structures generated in three dimensions. One option is to drive the network in such a way as to obtain two-dimensional growth, enabling a collection of independent images to be obtained. The density of the network generates ambiguous structures, such as overlaps and meetings, which hinder the reconstruction of the chronology of connections. In this paper, we propose a general method for global network reconstruction. Each network connection is defined by a unique label, enabling it to be tracked in time and space. In this work, we distinguish between lateral and apical branches on the one hand, and extremities on the other. Finally, we reconstruct the network after identifying and eliminating overlaps. This method is then applied to the model filamentous fungus Podospora anserina to reconstruct its growing thallus. We derive criteria for differentiating between apical and lateral branches. We find that the outer ring is favorably composed of apical branches, while densification within the network comes from lateral branches. From this, we derive the specific dynamics of each of the two types. Finally, in the absence of any latency phase during growth initiation, we can reconstruct a time based on the equality of apical and lateral branching collections. This makes it possible to directly compare the growth dynamics of different thalli., Comment: 30 pages, 11 figures

Context: Discrete symmetries have found numerous applications in photonics and quantum mechanics, but remain little studied in fluid mechanics, particularly in astrophysics. Aims: We aim to show how PT and anti-PT symmetries determine the behaviour of linear perturbations in a wide class of astrophysical problems. They set the location of Exceptional Points in the parameter space and the associated transitions to instability, and are associated to the conservation of quadratic quantities that can be determined explicitly. Methods: We study several classical local problems: the gravitational instability of isothermal spheres and thin discs, the Schwarzschild instability, the Rayleigh-B\'enard instability and acoustic waves in dust-gas mixtures. We calculate the locations and the order of the Exceptional Points with a method of resultant, as well as the conserved quantities in the different regions of the parameter space using Krein theory. Results: All problems studied here exhibit discrete symmetries, even though Hermiticity is broken by different physical processes (self-gravity, buoyancy, diffusion, drag). This analysis provides genuine explanations for certain instabilities, and for the existence of regions in the parameter space where waves do not propagate. Those correspond to breaking of PT and anti-PT symmetries respectively. Not all instabilities are associated to symmetry breaking (e.g. the Rayleigh-Benard instability)., Comment: 12 pages, 6 figures. Accepted for publication in A&A

We introduce in this paper a new formalisation of positive opetopes where faces are organised in a poset. Then we show that our definition is equivalent to that of positives opetopes as given by Marek Zawadowski.

We study two classes of radial integrals involving a product of bound and continuum one-electron states. Using a representation of the continuum part with an expansion on complex Gaussian Type Orbitals, such integrals can be performed analytically. We investigate the reliability of this scheme for low-energy physical parameters. This study serves as a premise in view of potential applications in molecular scattering processes., Comment: 19 pages, 4 figures

The Gaia Collaboration has recently reported the detection of a 33 M$_\odot$ black hole in a wide binary system located in the Solar neighbourhood. Here we explore the relationship between this black hole, known as Gaia BH3, and the nearby ED-2 halo stellar stream. We study the orbital characteristics of the Gaia BH3 binary and present measurements of the chemical abundances of ED-2 member stars derived from high-resolution spectra obtained with the VLT. We find that the Galactic orbit of the Gaia BH3 system and its metallicity are entirely consistent with being part of the ED-2 stream. The characteristics of the stream, particularly its negligible spread in metallicity and in other chemical elements as well as its single stellar population, suggest that it originated from a disrupted star cluster of low mass. Its age is comparable to that of the globular cluster M92 that has been estimated to be as old as the Universe. This is the first black hole unambiguously associated with a disrupted star cluster. We infer a plausible mass range for the cluster to be relatively narrow, between $2\times 10^3M_\odot$ and $4.2\times 10^4M_\odot$. This implies that the black hole could have formed directly from the collapse of a massive very-metal-poor star, but that the alternative scenario of binary interactions inside the cluster environment also deserves to be explored., Comment: Submitted to A&A Letters; updated author list

Gravitational waves from black-hole merging events have revealed a population of extra-galactic BHs residing in short-period binaries with masses that are higher than expected based on most stellar evolution models - and also higher than known stellar-origin black holes in our Galaxy. It has been proposed that those high-mass BHs are the remnants of massive metal-poor stars. Gaia astrometry is expected to uncover many Galactic wide-binary systems containing dormant BHs, which may not have been detected before. The study of this population will provide new information on the BH-mass distribution in binaries and shed light on their formation mechanisms and progenitors. As part of the validation efforts in preparation for the fourth Gaia data release (DR4), we analysed the preliminary astrometric binary solutions, obtained by the Gaia Non-Single Star pipeline, to verify their significance and to minimise false-detection rates in high-mass-function orbital solutions. The astrometric binary solution of one source, Gaia BH3, implies the presence of a 32.70 \pm 0.82 M\odot BH in a binary system with a period of 11.6 yr. Gaia radial velocities independently validate the astrometric orbit. Broad-band photometric and spectroscopic data show that the visible component is an old, very metal-poor giant of the Galactic halo, at a distance of 590 pc. The BH in the Gaia BH3 system is more massive than any other Galactic stellar-origin BH known thus far. The low metallicity of the star companion supports the scenario that metal-poor massive stars are progenitors of the high-mass BHs detected by gravitational-wave telescopes. The Galactic orbit of the system and its metallicity indicate that it might belong to the Sequoia halo substructure. Alternatively, and more plausibly, it could belong to the ED-2 stream, which likely originated from a globular cluster that had been disrupted by the Milky Way., Comment: 23 pages, accepted fro publication in A&A Letters. New version with small fixes

Topological properties of the spectrum of shallow-water waves on a rotating spherical body are established. Particular attention is paid to its spectral flow, i.e. the modes whose frequencies transit between the Rossby and inertia-gravity wavebands as the zonal wave number is varied. Organising the modes according to the number of zeros of their meridional velocity, we conclude that the net number of modes transiting between the shallow-water wavebands on the sphere is null, in contrast with the Matsuno spectrum. This difference can be explained by a miscount of zeros under the $\beta$-plane approximation. We corroborate this result with the analysis of Delplace et al (2017) by showing that the curved metric discloses a pair of degeneracy points in the Weyl symbol of the wave operator, non-existent under the $\beta$-plane approximation, each of them bearing a Chern number $-1$., Comment: 27 pages, 13 figures, 4 appendices

Neutral atom technology has steadily demonstrated significant theoretical and experimental advancements, positioning itself as a front-runner platform for running quantum algorithms. One unique advantage of this technology lies in the ability to reconfigure the geometry of the qubit register, from shot to shot. This unique feature makes possible the native embedding of graph-structured problems at the hardware level, with profound consequences for the resolution of complex optimization and machine learning tasks. By driving qubits, one can generate processed quantum states which retain graph complex properties. These states can then be leveraged to offer direct solutions to problems or as resources in hybrid quantum-classical schemes. In this paper, we review the advancements in quantum algorithms for graph problems running on neutral atom Quantum Processing Units (QPUs), and discuss recently introduced embedding and problem-solving techniques. In addition, we clarify ongoing advancements in hardware, with an emphasis on enhancing the scalability, controllability and computation repetition rate of neutral atom QPUs.

We implement and characterize a protocol that enables arbitrary local controls in a dipolar atom array, where the degree of freedom is encoded in a pair of Rydberg states. Our approach relies on a combination of local addressing beams and global microwave fields. Using this method, we directly prepare two different types of three-atom entangled states, including a W-state and a state exhibiting finite chirality. We verify the nature of the underlying entanglement by performing quantum state tomography. Finally, leveraging our ability to measure multi-basis, multi-body observables, we explore the adiabatic preparation of low-energy states in a frustrated geometry consisting of a pair of triangular plaquettes. By using local addressing to tune the symmetry of the initial state, we demonstrate the ability to prepare correlated states distinguished only by correlations of their chirality (a fundamentally six-body observable). Our protocol is generic, allowing for rotations on arbitrary subgroups of atoms within the array at arbitrary times during the experiment; this extends the scope of capabilities for quantum simulations of the dipolar XY model., Comment: 12 pages, 4 figures (main text) + 6 figures (Supplemental Material)

As programs to support efficient and sustainable energy sources are expanding, research into the potential applications of the hydrogen vector is accelerating. Proton exchange membrane fuel cells are electrochemical converters that transform the chemical energy of hydrogen into electrical energy. These devices are used today for low- and medium-power stationary applications and for mobility, in trains, cars, bicycles, etc. Proton exchange membrane fuel cells use a polymer membrane as the electrolyte. The role of the membrane is multiple: it must separate gases, be an electronic insulator and a very good ionic conductor. In addition, it must resist free-radical chemical attack and have good mechanical strength. Nafion-type perfluorinated membranes have all these properties: the fluorinated backbone is naturally hydrophobic, but the hydrophilic ionic groups give the material excellent water sorption properties. The water adsorbed in the structure is extremely mobile, acting as a transport medium for the protons generated at the anode. Although it has been studied for a long time and has been the subject of a large number of papers perfluorinated membranes are still the reference membranes today. This article reviews some contributions of Nuclear Magnetic Resonance methods in liquid state to the study of water properties in the structure of Nafion-type perfluorinated membranes.

Superconducting qubits have arisen as a leading technology platform for quantum computing which is on the verge of revolutionizing the world's calculation capacities. Nonetheless, the fabrication of computationally reliable qubit circuits requires increasing the quantum coherence lifetimes, which are predominantly limited by the dissipations of two-level system (TLS) defects present in the thin superconducting film and the adjacent dielectric regions. In this paper, we demonstrate the reduction of two-level system losses in three-dimensional superconducting radio frequency (SRF) niobium resonators by atomic layer deposition (ALD) of a 10 nm aluminum oxide Al2O3 thin films followed by a high vacuum (HV) heat treatment at 650 {\deg}C for few hours. By probing the effect of several heat treatments on Al2O3-coated niobium samples by X-ray photoelectron spectroscopy (XPS) plus scanning and conventional high resolution transmission electron microscopy (STEM/HRTEM) coupled with electron energy loss spectroscopy (EELS) and (EDX) , we witness a dissolution of niobium native oxides and the modification of the Al2O3-Nb interface, which correlates with the enhancement of the quality factor at low fields of two 1.3 GHz niobium cavities coated with 10 nm of Al2O3.

The Gaia satellite has already provided the astronomical community with three data releases, and the Radial Velocity Spectrometer (RVS) on board Gaia has provided the radial velocity for 33 million stars. When deriving the radial velocity from the RVS spectra, several stars are measured to have large values. To verify the credibility of these measurements, we selected some bright stars with the modulus of radial velocity in excess of 500\ to be observed with SOPHIE at OHP and UVES at VLT. This paper is devoted to investigating the chemical composition of the stars observed with UVES. We derived atmospheric parameters using Gaia photometry and parallaxes, and we performed a chemical analysis using the code. We find that the sample consists of metal-poor stars, although none have extremely low metallicities. The abundance patterns match what has been found in other samples of metal-poor stars selected irrespective of their radial velocities. We highlight the presence of three stars with low Cu and Zn abundances that are likely descendants of pair-instability supernovae. Two stars are apparently younger than 1\,Ga, and their masses exceed twice the turn-off mass of metal-poor populations. This makes it unlikely that they are blue stragglers because it would imply they formed from triple or multiple systems. We suggest instead that they are young metal-poor stars accreted from a dwarf galaxy. Finally, we find that the star RVS721 is associated with the Gjoll stream, which itself is associated with the Globular Cluster NGC\,3201., Comment: Astronomy and Astrophysics - A\&A, In press

We introduce a family of cluster algebras of infinite rank associated with root systems of type $A$, $D$, $E$. We show that suitable completions of these cluster algebras are isomorphic to the Grothendieck rings of the categories $\mathcal{O}_\mathbb{Z}$ of the corresponding shifted quantum affine algebras. The cluster variables of a class of distinguished initial seeds are certain formal power series defined by E. Frenkel and the second author, which satisfy a system of functional relations called $QQ$-system. We conjecture that all cluster monomials are classes of simple objects of $\mathcal{O}_\mathbb{Z}$. In the final section, we show that these cluster algebras contain infinitely many cluster subalgebras isomorphic to the coordinate ring of the open double Bruhat cell of the corresponding simple simply-connected algebraic group. This explains the similarity between $QQ$-system relations and certain generalized minor identities discovered by Fomin and Zelevinsky., Comment: 68 pages, 20 figures. V2 : references added, improved exposition in Sections 10.3, 10.4. V3 : 70 pages. Minor corrections and improvements. Final version to appear in PLMS

We argue that the spectrally cut-off Gaussian free field $\Phi_\Lambda$ on a compact Riemannian manifold or on $\mathbb{R}^n$ cannot satisfy the spatial Markov property. Moreover, when the manifold is reflection positive, we show that $\Phi_\Lambda$ fails to be reflection positive. We explain the difficulties one encounters when trying to deduce the reflection positivity property of the measure exp$(-\|\rho\Phi_\Lambda\|_{L^4}^4) \mu_{\text{GFF}}(d\Phi)$ from the reflection positivity property of the Gaussian free field measure $\mu_{\text{GFF}}$ in a naive way. These issues are probably well-known to experts of constructive quantum field theory but to our knowledge, no detailed account can be found in the litterature. Our pedagogical note aims to fill this small gap.

Using only 34 published tables, we reconstruct five variables (census block, sex, age, race, and ethnicity) in the confidential 2010 Census person records. Using the 38-bin age variable tabulated at the census block level, at most 20.1% of reconstructed records can differ from their confidential source on even a single value for these five variables. Using only published data, an attacker can verify that all records in 70% of all census blocks (97 million people) are perfectly reconstructed. The tabular publications in Summary File 1 thus have prohibited disclosure risk similar to the unreleased confidential microdata. Reidentification studies confirm that an attacker can, within blocks with perfect reconstruction accuracy, correctly infer the actual census response on race and ethnicity for 3.4 million vulnerable population uniques (persons with nonmodal characteristics) with 95% accuracy, the same precision as the confidential data achieve and far greater than statistical baselines. The flaw in the 2010 Census framework was the assumption that aggregation prevented accurate microdata reconstruction, justifying weaker disclosure limitation methods than were applied to 2010 Census public microdata. The framework used for 2020 Census publications defends against attacks that are based on reconstruction, as we also demonstrate here. Finally, we show that alternatives to the 2020 Census Disclosure Avoidance System with similar accuracy (enhanced swapping) also fail to protect confidentiality, and those that partially defend against reconstruction attacks (incomplete suppression implementations) destroy the primary statutory use case: data for redistricting all legislatures in the country in compliance with the 1965 Voting Rights Act.

The ionization by photon or electron impact of the inner (2a1) and outer (1t2) valence orbitals of the CH4 molecule is investigated theoretically. In spite of a number of approximations, including a monocentric approach and a rather simple distorting molecular potential, the calculated cross sections are overall similar to those of other theoretical methods, and in reasonable agreement with experimental data. The originality of the present approach stands in the way we evaluate the transition matrix elements. The key ingredient of the calculation scheme is that the continuum radial wave function of the ejected electron is represented by a finite sum of complex Gaussian type orbitals. This numerically expensive optimization task is then largely compensated by rather simple and rapid analytical calculations of the necessary integrals, and thus all ionization observables, including cross section angular distributions. The proposed and implemented Gaussian approach is proved to be numerically very reliable in all considered kinematical situations with ejected electrons up to 2.7 a.u.. The analytical formulation of the scheme is provided here for bound molecular states described by monocentric Slater type orbitals; however, Gaussian type orbitals may be equally be employed. In combination with complex Gaussian functions for the continuum states, an all Gaussian approach with multicentric bound states can be envisaged.

In "The 2020 Census Disclosure Avoidance System TopDown Algorithm," Abowd et al. (2022) describe the concepts and methods used by the Disclosure Avoidance System (DAS) to produce formally private output in support of the 2020 Census data product releases, with a particular focus on the DAS implementation that was used to create the 2020 Census Redistricting Data (P.L. 94-171) Summary File. In this paper we describe the updates to the DAS that were required to release the Demographic and Housing Characteristics (DHC) File, which provides more granular tables than other data products, such as the Redistricting Data Summary File. We also describe the final configuration parameters used for the production DHC DAS implementation, as well as subsequent experimental data products to facilitate development of tools that provide confidence intervals for confidential 2020 Census tabulations.