Machine Learning
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Tuesday 8 September
In this paper we derive a Probably Approximately Correct (PAC)-Bayesian error bound for partially observed linear time-invariant (LTI) stochastic dynamical systems in state-space form with inputs and sub-Gaussian noise. Such bounds are widespread in machine learning, and they are useful for characterizing the predictive power of models learned from finitely many data points. The bound derived in this paper relates the expectation of prediction errors with the prediction error generated by the model on the data used for learning. In addition, we show that it can also be used to derive bounds for the parameter estimation error. In turn, this allows us to provide finite-sample error bounds for the prediction error and parameter estimation error for a wide class of system identification algorithms. Furthermore, as LTI systems are a sub-class of recurrent neural networks (RNNs), these error bounds could be a first step towards PAC-Bayesian bounds for RNNs.
Distributed Computing
Tools-CC-Bench: a Benchmark Suite for Collective Communication with Compression in HPC and AI Workloads
Distributed HPC and LLM workloads increasingly require efficient communication for scalability, yet growing data movement has become a major performance bottleneck. Communication compression can reduce this overhead and complement execution-level optimizations, but its benefits remain difficult to assess because existing benchmarks lack support for diverse backends, realistic datasets, application-specific accuracy metrics, and overlap-induced resource contention. We present CC-Bench, a lightweight, extensible, and application-oriented benchmark suite for evaluating communication compression under realistic execution conditions. CC-Bench uses declarative application-environment modeling to decouple profiling logic from communication libraries, datasets, and fidelity metrics, enabling portable cross-library evaluation. It further combines function-level interception and hardware counter monitoring to characterize per-phase latency, hardware utilization, numerical fidelity, and computation interference. With representative datasets from HPC and LLM workloads, CC-Bench evaluates three compression-enabled communication libraries on CPU and GPU clusters, revealing accuracy-performance trade-offs and bottlenecks to guide practical deployment and optimization.
Read the full abstract
Analysis of PDEs
Transport and flow for horizontal Sobolev contact velocities in Carnot groups
We establish new well-posedness results for transport and flow equations driven by contact vector fields on Carnot groups. The velocity fields are assumed to have horizontal Sobolev regularity, namely Sobolev regularity only along the horizontal directions determined by the stratified geometry of the group. In the broader sub-Riemannian setting, results of this type were previously known only for Heisenberg groups. Our proof relies on the theory of renormalized solutions as originally introduced by DiPerna and Lions in the Euclidean setting.
Read the full abstract
High Energy Astrophysical Phenomena
A Unified Timescale Relation for Quasi-Periodic Eruptions and Repeated Nuclear Transients
Quasi-periodic eruptions (QPEs) and recurrent nuclear transients (RNTs) exhibit recurrent high-amplitude flares from galactic nuclei, yet their characteristic timescales remain poorly understood. In this work, we compile a sample of these systems and investigate empirical scaling relations between flare timescales and black hole masses. We find that the recurrence timescale exhibits a positive but highly scattered dependence on black hole mass in the combined QPE and RNT sample, approximately following $t_{\rm rec}\propto M_{\rm BH}^{1.19^{+0.52}_{-0.47}}$ with an intrinsic scatter of 0.95 dex. Remarkably, we uncover a tight nearly linear relation between recurrence time and flare timescale for QPEs and RNTs, described by $t_{\rm rec}\propto t_{\rm rise}^{1.00\pm0.07}$ with an intrinsic scatter of 0.29 dex. This relation extends across timescales from hours for QPEs to months and years for nuclear transients. We further find that QPEs and RNTs approximately follow a common empirical relation between recurrence time and flare rise time, although the physical origin of this relation remains uncertain. Our findings reveal a common phenomenological timescale link across RNTs, providing a practical framework for characterizing their temporal behavior.
Read the full abstract
Artificial Intelligence
When Can One Obtain Certificates of Optimality Using Positivstellensaetze?
We study certificates of positivity and optimality for learning problems whose objectives and constraints need not be polynomial. We isolate an axiomatic core of Fischer's constructive strict and weak Positivstellensätze and prove the resulting theorems for abstract function algebras over ordered fields. The framework separates two roles that can otherwise be conflated: objective and constraint functions may be built from broad classes of continuous or definable operations, while the auxiliary primitives used to construct a certificate satisfy explicit scalar and closure axioms. We give instances over continuous and definable function algebras, including ordered fields not closed under square roots, derive lower-bound and global-optimality certificates, and analyze both expanded term length and shared computation-graph complexity.
Read the full abstract
Combinatorics
Orthogonal signed graphs of degree 5
An orthogonal signed graph is a connected signed graph whose signed adjacency matrix has pairwise orthogonal rows. They are closely related to Hadamard matrices, maximal arrangements of equiangular lines, bipartite Ramanujan graphs, and the remarkable resolution of the Sensitivity Conjecture. Orthogonal signed graphs of degree at most 4 have been completely classified, and partial results were known for degree 5. We complete the classification of orthogonal signed graphs of maximum degree 5. We also provide several new infinite families of 6- and 8-regular orthogonal signed graphs.
Read the full abstract
Materials Science
Study on Thickness and Temperature Dependence of Thermoelectric Properties in SnS Nanofilms
SnS as an environmentally friendly, cost-effective, and earth-abundant narrow-bandgap semiconductor material, has demonstrated significant application potential in the field of medium-temperature thermoelectric conversion. However, the thermoelectric performance of its bulk counterpart is inherently constrained by intrinsic point defects (e.g., vacancies) and the material's specific band structure. Low-dimensional engineering has emerged as a pivotal strategy for overcoming these limitations and enhancing thermoelectric performance. In this work, we systematically investigate the thermoelectric properties of SnS nanofilms with distinct thicknesses (82 nm, 199 nm, 616 nm, and 813 nm) across a temperature range of 300-600 K. Measurements were conducted using time-domain thermoreflectance (TDTR) and a dedicated thin-film thermoelectric parameter test system (ZEM-3). Our results confirm that low-dimensionalization effectively boosts the thermoelectric performance of SnS, with the thermoelectric figure of merit (ZT) displaying a pronounced dependence on both film thickness and temperature. All four SnS thin films exhibit thermoelectric performance that is markedly superior to that of bulk SnS. This enhancement is primarily attributed to the quantum confinement effect, energy filtering effect, and intensified phonon scattering, all of which are induced by the low-dimensional structural characteristics. This work provides not only experimental evidence and theoretical insights for the performance optimization of SnS nanofilms but also establishes a foundational framework for the development of high-efficiency, eco-friendly medium-temperature thermoelectric materials, thereby holding significant scientific value and practical implications.
Read the full abstract
Quantum Physics
On the Power of Adaptivity in Testing Quantum States in Fidelity
We study the problems of quantum state certification, equivalence testing and independence testing. In certification, given samples of an unknown quantum state $\rho$ and the description of a state $\sigma$, the goal is to test whether $\rho=\sigma$, or whether $\rho$ and $\sigma$ are far in a given distance measure. In equivalence testing, $\sigma$ is also unknown and only accessible via samples. Independence testing decides whether $\rho_{AC}=\rho_A\otimes\rho_C$, or is far from being a product. The sample complexities of these problems are now well-understood for a decision gap $\varepsilon$ in trace distance: in the single-copy measurement setting with $d$-dimensional states, all three tasks can be solved using the same non-adaptive approach, which uses $\Theta(d^{3/2}/\varepsilon^2)$ samples and is optimal in general, even without adaptivity. In this work, we consider decision gaps expressed in fidelity and study possible separations between these problems and how adaptivity can help. We prove that certification with respect to fidelity for a state $\sigma$ of rank $r$ does not benefit from adaptivity and requires $\widetilde{\Theta}(r^{3/2}/\varepsilon)$ samples. For equivalence testing and independence testing, we provide adaptive algorithms using $\widetilde{O}(\min\{d^{3/2}/\varepsilon^2,d^{9/4}/\varepsilon\})$ and $\widetilde{O}(\min\{(d_Ad_C)^{3/2}/\varepsilon^2,d_A^{9/4}d_C^{3/4}/\varepsilon\})$ samples, for $d_A\geq d_C$, respectively. Our main technique is a framework that uses partial learning and a reduction to testing in $\ell_2$-distance, adapted from the distribution testing literature. We show that adaptivity matters for equivalence testing in fidelity by proving that $\widetilde{\Omega}(1/\varepsilon^2)$ samples are necessary in the non-adaptive case even for qubits, showing a separation from certification.
Read the full abstract
Analysis of PDEs
Shape optimization of light structures with general Hooke's laws
For the minimum-compliance shape optimization problem it was established recently [Duke Math.\ J. 172 (2023), 43–103] that a sequence of approximately optimal shapes of exact volume \(\e\) converges to a limiting generalized shape, which is represented by a possibly diffuse probability measure, as \(\e \todown 0\). This generalized shape minimizes the so-called light-structures compliance functional, which — as conjectured by Bouchitté — is expressed in terms of a relaxed integrand, reflecting that optimal shapes must only involve “lower-dimensional” structures, like the Michell-type trusses that are well known in the engineering domain. However, the aforementioned result only applies for the simplest isotropic quadratic energy density, that is, the squared Frobenius norm, in dimensions two and three. The present work extends the relaxation result to the entire class of (isotropic or anisotropic) elasticity tensors and to any dimension, thus solving Bouchitté's conjecture in full generality. Since the previous work relied on explicit formulas for the relaxed integrands (due to Allaire, Kohn, and Strang), which are not available in the general case, several new arguments and substantial modifications of the original approach are required. We also significantly simplify and streamline other parts of the original argument, in addition to clarifying some related statements in the literature.
Read the full abstract
High Energy Astrophysical Phenomena
Stratified Wind Geometries of GRS 1915+105: Density Profiles and Radii
The low-mass X-ray binary GRS 1915+105 exhibits dramatic spectral and flux variability, often accompanied by highly ionized outflows. In this work, we analyze four archival Chandra/HETG observations spanning over a decade, which vary greatly in their ionizing spectrum from soft (low inner disk temperature) to hard (high inner disk temperature). We measure the wind properties, using the insightful tool of the absorption measure distribution (AMD, a distribution of column density with ionization). While the soft epoch reveals a steep, monotonically rising AMD, hard epochs feature a distinct turnover at high ionization parameter ($\log\xi>3.3$). This is the first time we observe a negative AMD slope in an outflow, which we interpret as a localized density inversion, which could point to radiation-pressure compression zones. We utilize the AMD shape and total column density measurements to calculate the density profile and distance of the wind from the central source. For the hard states, we obtain densities of $10^{12}-10^{13}$ cm$^{-3}$ at radii of $10^{10}-10^{11}$ cm. Such measurements are usually not directly possible in X-ray spectroscopy. Here we constrain the inner radius and density of the outflow to a factor of two. The high-$\xi$ values up to $\log\xi\sim6.5$ with no turnover in the soft state imply a smaller radius of a few $10^{8}$cm, which is in tension with the slow velocities observed. We demonstrate that this is a generic tension between compact launch radii and sub-escape velocities of high-column, high-ionization outflows observed in stellar-mass black hole binaries.
Read the full abstract
Computer Vision
CVT-GS: Learning to Simplify 3D Gaussian Splatting with Centroidal Voronoi Tessellation
While 3D Gaussian Splatting (3DGS) has emerged as a powerful representation for real-time novel view synthesis, rendering high-fidelity scenes often relies on a massive number of Gaussian primitives, incurring substantial storage and computational overhead. Existing simplification techniques are largely intrusive, requiring training-time pruning, architectural modifications, or computationally expensive per-scene fine-tuning. These drawbacks limit their deployment on off-the-shelf pretrained models. In this paper, we propose CVT-GS, a novel optimization-free post-hoc simplification framework that directly compresses trained 3DGS scenes without sacrificing visual fidelity. Our approach first constructs spatially coherent cells over Gaussian centers via a geometry-aware Centroidal Voronoi Tessellation (CVT). Subsequently, a lightweight neural cell merger predicts the geometry and appearance of a single, highly representative Gaussian primitive for each cell under differentiable rendering supervision. By formulating simplification as a rendering-aware many-to-one merging process rather than naive primitive pruning, CVT-GS outputs a standard 3DGS scene that is seamlessly compatible with existing renderers. Experiments on various datasets demonstrate the superiority of our method. Notably, when achieving a 100-fold reduction in Gaussian points, our method operates 12 times faster than state-of-the-art methods while improving the PSNR by 1.3 dB.
Read the full abstract
Artificial Intelligence
Application of curiosity driven exploration methods for hardware interference identification
The transition from single-core to multi-core architectures in safety-critical embedded systems introduces significant challenges due to inter-core interference caused by contention for shared hardware resources. Such interference affects execution times and complicates the verification of strict temporal requirements, particularly in domains such as avionics where standards require comprehensive identification of interference sources. Existing interference analysis approaches, whether manual or model-based, struggle to capture the full range of behaviors arising from the complex interactions among micro-architectural components. In this paper, we frame multi-core interference analysis as the exploration of a complex system behavior space. We propose the use of curiosity-driven exploration algorithms from artificial intelligence to systematically and efficiently cover the space of possible interference behaviors. Using a simulator-based environment, we show that the proposed approach achieves broader and more uniform behavioral coverage within a limited experimental budget compared to traditional pseudo-random program generation methods.
Read the full abstract
Number Theory
Generalizing a Pair of Diophantine Equations
For coprime integers $a$ and $b$, it is known that exactly one of the two Diophantine equations $$ ax+by\ =\ \frac{(a-1)(b-1)}{2} \qquad\text{and}\qquad 1+ax+by\ =\ \frac{(a-1)(b-1)}{2} $$ admits a nonnegative integer solution, and that this solution is unique. We first generalize this result by replacing the right-hand side with an arbitrary integer $m$ and its complement $ab-a-b-m$. This framework enables us to study the existence and uniqueness of nonnegative integer solutions to $$ ax+by\ =\ \frac{(a-1)(b-1)}{k} \qquad\text{and}\qquad 1+ax+by\ =\ \frac{(a-1)(b-1)}{k}, $$ where $k$ is a fixed positive integer. We then obtain explicit results when $a$ and $b$ are consecutive Fibonacci numbers. Finally, we examine the original pair of equations in several particular settings, including when $b\equiv \pm1\mod a$, when $b$ is replaced by a higher power, and when the parameters are squared.
Read the full abstract
Quantum Physics
Adaptive AI for Pulse-Level Quantum Control
The Control Variational Quantum Eigensolver (ctrl-VQE) directly optimizes microwave pulses to enable faster and lower-error quantum-state preparation, but its continuous control landscape re- quires efficient search strategies. We demonstrate that a reinforcement-learning agent based on a deep Q learning network can autonomously discover high-performance pulse sequences using only system parameters and a reward function. The approach is fully general for superconducting qubit platforms, requires no ansatz, and operates at nanosecond resolution compatible with hardware con- straints. As a proof of concept, we apply the method to ground-state preparation of the Hydrogen molecule on a simulated superconducting device. The agent consistently identifies optimized control sequences that achieve high fidelity and outperform random-search baselines. These results highlight adaptive learning as a promising hardware-ready framework for pulse-level quantum control.
Read the full abstract
Probability
Asymptotics of a planar isoperimetric problem with a white-noise volume term
In this work we study the maximal ratio $I$ between the white noise integrated over a set and the perimeter of the set, which can be seen as a random isoperimetric problem, and appears in the random-field Ising model (Ding-Wirth) and min-max optimal matching (Leighton-Shor). In the planar case considered here, such a ratio is scale-invariant and therefore the problem is critical. As such it requires an ultraviolet cutoff, which we impose by restricting to polygonal sets with side-length at least 1 contained in the ball $B_L$. Our main result establishes the leading-order asymptotics $\mathbb{E}I\approx i\ln^{3/4}L$ for some $i\in(0,\infty)$ and superconcentration at scale $O(\ln^{-1/4}L)$. This is done by relating to a simpler $(1+1)$-dimensional action, studied by the last two authors and C. Wagner. Such a connection is found by a classical geometric linearization of the perimeter to the Dirichlet energy, justified by Ried-Wagner large-scale regularity theory, and allows a coarse-graining and scale-by-scale iteration argument.
Read the full abstract
Machine Learning
BAFF: Bid-Aware Filter Family for Mitigating Training Data Interference in RTB A/B Tests
In online A/B tests for real-time bidding (RTB), control and treatment models are typically trained on a shared serving log that includes data generated by the counterpart model. This shared-log training biases each model's training data through two channels: the counterpart model may have selected a different ad from the ad-candidate pool (ad-ranking disagreement) and may have bid a different price (bid-pricing disagreement), potentially distorting the A/B test outcome. Log-splitting eliminates the bias but sacrifices training data; log-sharing retains all data but leaves the bias unaddressed. We formalize the Bid-Aware Filter Family (BAFF), a class of (k,l)-parameterized hard filters that controls tolerance to each channel independently, providing a structured search space between these two extremes. We further propose a three-stage online measurement protocol that enables evaluating data-sharing strategies by their deviation from an interference-free reference model in production. In offline simulation, a (k,l) sweep surfaces operating points with smaller deviation from the interference-free reference model than both log-sharing and log-splitting. In a live RTB deployment on a demand-side platform (DSP), filter-based variants preserve the reference model's business metrics (e.g., CPC, CTR) more closely than both baselines. The best operating point is setting-dependent, underscoring the practical value of the search space itself.
Read the full abstract
Number Theory
On the cardinality of the exception set in Littlewood's discrete conjecture
Let $p$ be a prime, $d\ge 2$, $H\in [1,p)$, and $\ln\ln p = o(\ln H)$. We prove that $$ \min_{1\le n \le H} n \left\|\frac{a_1 n}{p}\right\|\ldots \left\|\frac{a_d n}{p}\right\| \approx \frac{1}{(\ln p)^{d-1} \ln H} $$ for "almost all" $a \in ({\Bbb Z} / p{\Bbb Z})^d$.
Systems and Control
NL-FRA: A MATLAB package for nonlinear frequency response analysis
Nonlinear Output Frequency Response Functions (NOFRFs) provide a one-dimensional frequency-domain representation of nonlinear dynamics, enabling direct decomposition of an output spectrum into contributions from different orders of nonlinearity. NL-FRA (NonLinear Frequency Response Analysis) is an open-source MATLAB package implementing a data-driven Least Squares (LS) method for estimating NOFRFs directly from system input-output data. It provides an integrated workflow for NOFRF estimation, validation, and visualisation, together with nonlinear transmissibility analysis. The package supports different types of inputs, i.e. general band-limited, multi-one and harmonic inputs, and includes routines for identifying the frequency ranges over which each nonlinear order contributes to the output. Furthermore, since the LS approach estimates NOFRFs directly from input-output data, it can be applied to experimental measurements or data generated using any suitable dynamic model of the system. The software facilitates interpretable nonlinear frequency-domain analysis for applications including fault diagnosis, condition and structural health monitoring, and biomedical engineering.
Read the full abstract
Graphics
TBR: Transport-Based Rendering with Deposition Strokes for Inverse Graphics
We present a stroke design in which strokes are transport-coupled: each stroke deposits material of its own area and moves every earlier mark without changing its area, so later strokes deform earlier ones. We then solve the inverse problem under this design: given a target image, we optimise an ordered program of such strokes whose replay approximates it, with digital marbling as the motivating medium. The stroke is a capsule that continuously joins circular drops to drawn deposits; its transport is exactly area-preserving, with a closed-form inverse outside the deposit, and a variant with the same inverse differs from line-source potential flow by 8% of the mean displacement. A replay adjoint regenerates intermediate states instead of storing them and uses 8.7x less memory than checkpointed automatic differentiation; a fused implementation fits a 2000-stroke program at 1024x1024 in about four minutes on one GPU. On five marbled sheets the recovered programs are level with a published stroke-based fitter as rasters, replay across a fourfold resolution range, and support edits in program order and palette space that stay valid under transport.
Read the full abstract
Chemical Physics
Competing Ring-Opening and Hofmann Elimination Pathways in Aqueous TEMPO Catholytes: A First-Principles Study
Aqueous redox-flow batteries based on TEMPO derivatives are promising for large-scale energy storage, but their practical use is limited by the chemical instability of the oxidized N -oxoammonium state. In this work, we investigate the degradation of five TEMPO derivatives using ab initio molecular dynamics combined with enhanced sampling. Two proposed degradation mechanisms, ring opening and Hofmann elimination, are examined and their corresponding activation free energies are compared. For all derivatives considered, ring opening exhibits a lower activation free energy than Hofmann elimination, identifying it as the kinetically preferred degradation pathway. The magnitude of the ring-opening barrier, however, varies significantly between molecules, showing that different functionalizations strongly influence its stability toward degradation. The predicted preference for ring opening is consistent with available experimental studies, which have identified or inferred ring-opening degradation for several TEMPO-based catholytes. These results provide an atomistic picture of degradation pathways that are difficult to resolve experimentally and highlight the importance of molecular structure in controlling the kinetic stability of TEMPO derivatives in aqueous electrolytes.
Read the full abstract
Differential Geometry
Dirac eigenvalues and spacetime mean curvature estimates for DEC spin fill-ins
We establish spacetime counterparts of fill-in inequalities for spin initial data sets satisfying the dominant energy condition in dimensions three through seven, controlling the minimum and the integral of the norm of the spacetime mean curvature vector by intrinsic boundary data. The proof relies on a Reilly–MIT estimate for complete spin manifolds with compact boundary. As further consequences, under suitable nonnegativity assumptions, Gromov's hyperspherical-radius fill-in inequality and Bär's recent total mean-curvature estimate extend to complete, possibly noncompact spin manifolds with compact boundary.
Read the full abstract
Artificial Intelligence
GoAnt: Quality-Diversity Multi-Agent Search for Alpha Factor Discovery in Market Microstructure Data
Automated alpha factor discovery searches symbolic trading signals from price-volume panels and order-book data under a fixed evaluation budget. Existing single- and multi-agent program-search systems can overfit predictive proxies that fail after execution costs and repeatedly explore redundant factor families, limiting execution robustness and behavioral diversity. We introduce GoAnt, a quality-diversity multi-agent search framework that combines non-communicating Explorer, Exploiter and Connector workers with a shared adaptive Mental Map and a compact Queen dispatcher. The Mental Map organizes candidates by leakage-free execution profiles and retains one elite per niche, while the Queen reallocates the evaluation budget from explicit search-state summaries. We also define a map-independent effective-yield protocol that counts high-quality, mutually nonredundant factors directly from each method's evaluation records, giving archive-based and map-free systems the same ruler. On real A-share microstructure data spanning 2023–2026, GoAnt reaches quality-weighted yields of 41.8 and 47.6 in price-volume and order-book settings, improving the strongest baseline by 57% and 97% under matched budgets. Its locked populations retain 0.64 and 0.67 of in-sample quality out of sample, compared with 0.61 and 0.63 for a static map.
Read the full abstract
Probability
Sharp regularity and small ball probabilities for the stochastic heat equation on bounded domains
We consider the stochastic heat equation $\partial_t u(t,x) = \Delta u(t,x) + \dot{W}_\alpha(t,x)$ on a bounded Lipschitz domain with zero Dirichlet boundary condition and zero initial condition, where $\dot{W}_\alpha$ is a Gaussian noise that is white in time and whose spatial covariance is the kernel of $(-\Delta)^{-\alpha}$ with $\alpha>0$. We prove that a unique pointwise defined mild solution exists if and only if $\alpha>d/2-1$. In this case, if in addition the domain is $C^2$, we also establish spatial and temporal Holder regularity of the solution. When $d/2-1<\alpha<d/2$, we show that the Holder exponents are optimal and obtain exact local and uniform moduli of continuity, a Chung-type law of the iterated logarithm, and sharp small ball probability estimates for the solution.
Read the full abstract
Probability
Exact Asymptotics for the 2D Euclidean Random Matching Problem
We determine the exact first-order asymptotics of the expected optimal cost in two-dimensional random bipartite matching, for every finite power cost $q \ge 1$, on the flat torus. In the endpoint case $q=1$, this answers a question by Talagrand, in the periodic case. The argument involves the closely related asymptotics of the energy of the solution of the $p$-Poisson equation with a regularized white-noise source. In the limit of vanishing regularization parameter, we identify this energy as the solution of a Variational Martingale Problem on a limiting Gaussian filtration, whose value is characterized by a parabolic Monge-Ampère flow.
Read the full abstract
Quantum Physics
Tracing the Loop: Non-Causal Computation, Partial Traces, & Postselected Entanglement
This paper gives a categorical interpretation of Baumeler & Wolf's logically consistent non-causal circuits, connecting them to postselected quantum teleportation. Looped feedback is represented by a trace in the category of non-negative matrices, and it is shown that the traced process is stochastic precisely when the induced loop transition matrix has trace $1$ for every external input, a condition shown to be equivalent to a unique fixed point for the loop for each input to a deterministic circuit. A classical non-causal circuit is represented by a measure-and-prepare quantum channel with an internal register utilising a maximally entangled Bell-state with postselection. The main result is that classical logical consistency is equivalent to the postselected Bell outcome having, for loop dimension $d$, probability exactly $1/d^2$ for each classical input distribution. The subsequent normalised conditional output then agrees exactly with the classical categorical trace. This work identifies a class of quantum Bell-postselection constructions whose conditional evolution maintains linear dependence on classical input distributions.
Read the full abstract
Analysis of PDEs
The $\mathrm{L}^1$-Stokes Semigroup
We study the Stokes operator with no-slip boundary conditions on the spaces $\mathrm{L}_{\sigma,n}(\Omega)$ and ${\mathrm{L}^1(\Omega,\mathbb{C}^d)}/{\nabla \mathrm{W}^{1,1}(\Omega,\mathbb{C})}$, where $\Omega\subset\mathbb{R}^d$ is an arbitrary bounded $\mathrm{C}^{1,\alpha}$-domain. We show that the Stokes operator on $\mathrm{L}^1_{\sigma,n}(\Omega)$ does not generate a $\mathrm{C}_0$-semigroup, even though the resolvent problem is uniquely solvable. In stark contrast, its realization on ${\mathrm{L}^1(\Omega,\mathbb{C}^d)}/{\nabla \mathrm{W}^{1,1}(\Omega,\mathbb{C})}$ generates a compact, analytic $\mathrm{C}_0$-semigroup, which leaves $\mathrm{L}^1_{\sigma,n}(\Omega)$ invariant. The key point is that these two realizations, which are canonically identified for $1<p<\infty$ through the Helmholtz decomposition, cease to be equivalent at the endpoint $p=1$. This leads to genuinely different functional analytic properties. Our result provides the first positive generation theorem for the Stokes operator with no-slip boundary conditions in a pure $\mathrm{L}^1$-setting on a bounded domain and settles a problem that had remained open for nearly fifty years; see, e.g., \cite{Koz:01,DHP:01}. In this sense, it completes the theory of the Stokes semigroup across the full scale of solenoidal Lebesgue spaces on (smooth) bounded domains. As an intermediate step, some results on the space of Radon measures are obtained. The proof combines the sun-dual construction with a precise analysis of the failure of the Helmholtz decomposition in $\mathrm{L}^1$, the celebrated result of Abe and Giga \cite{AG:12} on the Stokes semigroup on $\mathrm{C}_{\sigma,0}(\Omega)$ and the regularity theory refinements for the Stokes operator recently developed by Breit and the second author \cite{BG:25}.
Read the full abstract
Probability
A large-deviation principle for the empirical distribution of a regular branching random walk
Let \(\{Z_n\}_{n\geq0}\) be a supercritical branching random walk with deterministic rooted \(b\)-ary tree, \(b\geq2\), and symmetric displacements satisfying \(\lim_{x\to+\infty}x^{-\alpha}\log\Pp(X>x)=-\lambda \) with \(\alpha,\lambda>0\). Set $\overline Z_n(\cdot):=Z_n(\cdot)/Z_n(\mathbb{R}),~n\geq0.$ For a finite union \(A\) of intervals, we establish the following full large-deviation principle: for every Borel set \(\Gamma\subset[0,1]\), \begin{align*} -\inf_{q\in\Gamma^\circ}Q_A(q) &\leq \liminf_{n\to\infty}n^{-\alpha/2} \log\Pp\!\left(\overline Z_n(\sqrt n\,A)\in\Gamma\right) &\leq \limsup_{n\to\infty}n^{-\alpha/2} \log\Pp\!\left(\overline Z_n(\sqrt n\,A)\in\Gamma\right) &\leq -\inf_{q\in\overline\Gamma}Q_A(q), \end{align*} where $Q_A$ is a good rate function on $[0,1]$. Meanwhile, we obtain large deviation probabilities for $\{\overline Z_n(\sqrt n\,A)\}_{n\geq1}.$ This strengthens the nonmatching upper and lower bounds obtained by Chen and He [Probab. Theory Related Fields 175 (2019) 255-307] for regular trees with Weibull displacements. Our method combines a large-deviation principle for rescaled displacement tree fields and exponential equivalence.
Read the full abstract
Combinatorics
Coloring Small $K_t$-Minor-Free Graphs
Delcourt and Postle proved that every $K_t$-minor-free graph is $O(t\log\log t)$-colorable and reduced the Linear Hadwiger Conjecture to coloring $K_t$-minor-free graphs on $O(t\log^4 t)$ vertices. In this paper, we use the star contraction technique to improve their bound for small graphs and use their reduction to extend this improvement to all $K_t$-minor-free graphs. Thus we improve their $O(t\log\log t)$ bound to $O(t\sqrt{\log\log t})$.
Read the full abstract
High Energy Physics (Phenomenology)
From LUX-ZEPLIN to Colliders: Probing Higgsino Dark Matter
The high-energy nuclear recoil event with recoil energy $E_R \approx 248\text{ keV}$ observed by the LZ collaboration provides an exciting hint toward a model with a $1.1\text{ TeV}$ Higgsino inelastic dark matter. Such a recoil energy requires a mass splitting of order ($\delta \approx 350\text{ keV}$) between the two nearly-degenerate neutral states. We show that within the framework of the MSSM, the model predicts a nearly-degenerate charged Higgsino state, chargino, whose mass splitting from the neutral states is of order $\mathcal{O}(350)$~MeV. The subsequent decays of such charginos once produced at colliders would lead to a sub-centimeter charged track or tracklet (with lifetime $\tau \approx 0.025\text{ ns}$) at the detector. We show that such a scenario is not constrained by the current LHC bounds, and could be tested at future high-energy colliders, including HL-LHC, 100 TeV $pp$ colliders, and muon colliders. The most critical requirement is how short a track or tracklet can be reconstructed.
Read the full abstract
Robotics
DCLP++: Learning to Navigate with Footprint Clearance and Relative Motion
We present DCLP++, a local navigation frameworkthat uses footprint clearance as the geometric basis for studying relative motion features in dynamic environments. Each valid LiDAR return is mapped to its shortest Euclidean distance from the filled robot footprint before reciprocal encoding, replacing distance from the sensor with distance to the occupied body. Radial measurementsor simulated planar relative velocities provide short-horizon features without static-dynamic labels in the policy input. A preliminary study uses a rectangular robot with a speed limit of 1 m/s among 20 moving obstacles. On 100 fixed validation tasks, two selected training seeds yield mean success rates of 42% with sensor rangeand 70% with footprint clearance after 200,000 environment steps.Motion variants show mixed additional gains. These results supportthe clearance-based observation in the evaluated setting; reliable motion benefits and transfer across robots require further evaluation.