Peer Revenue Now accepting positions

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

All of arXiv

Monday 9 November

Algebraic Geometry

Motivic Mirror Symmetry for Sink-recurrent Cluster Varieties

We prove motivic mirror symmetry for full-rank sink-recurrent cluster varieties. We allow skew-symmetrizable principal parts and invert all frozen variables. The rational geometric motive of such a variety is isomorphic to the additive orbifold motive of the mirror stack of Hu and Shende. This gives comparisons of cohomology, higher Chow groups, and algebraic $K$-theory with rational coefficients. The proof combines quantum deformation with the splitting of these motives into Tate summands. For really full-rank skew-symmetric seeds, a finite étale morphism realizes the comparison. We also study this geometric comparison in families over the projective line.
Read the full abstract

Analysis of PDEs

Quantitative global regularity for Navier–Stokes equations with odd viscosity on the cubic three-torus

We study the three-dimensional incompressible Navier–Stokes equation on the cubic torus with a constant odd-viscosity operator given by the vertical derivative of the curl. Arbitrary real, mean-zero, divergence-free initial data and an independent steady force, both in the second-order Sobolev space, admit a unique global strong solution when the magnitude of one fixed odd-viscosity coefficient is at least a norm-uniform sufficient threshold. In the unforced case the solution is smooth at positive times. The qualitative argument adapts the resonant-averaging framework for rapidly rotating fluids to the different odd-viscous dispersion. Weighted lattice counts and signed corrected-energy estimates yield a sufficient bound for the odd-to-ordinary viscosity ratio that is polynomial in the normalized initial-data radius at each fixed normalized force radius, together with a separate bound in a specified force-dominated regime. No optimality is claimed for the polynomial exponents or prefactors; determining the sharp viscosity threshold for the original equation remains open. Direct kinetic and weighted enstrophy barriers also give regularity for every coefficient on explicit bounded data and force regions, including the closed unforced ball of homogeneous second-order Fourier Sobolev radius ninety-seven fortieths times the viscosity and seven closed forced rectangles. We specify how finite certificates enter the analytic argument and distinguish minimization of a sufficient formula from sharpness for the original equation. All three main theorems are formally verified in Lean 4 with Mathlib.
Read the full abstract

Cosmology and Nongalactic Astrophysics

Little Red Dots from Primordial Black Holes: Non-Gaussianity and Accretion Evade $\mu$-Distortion Constraints

Recent JWST observations of "Little Red Dots" (LRDs) at $z \gtrsim 4$ suggest the existence of early supermassive black holes (SMBHs) with masses $\gtrsim 10^6\,M_\odot$. Their rapid emergence challenges conventional seeding and growth scenarios based on accretion and mergers of Population III stellar remnants, motivating renewed interest in primordial black holes (PBHs) formed from primordial perturbations. Explaining such massive seeds requires power spectrum enhancements that are constrained by cosmic microwave background (CMB) spectral distortions. In this work we perform the first comprehensive Bayesian exploration of non-Gaussian primordial perturbations capable of producing SMBH seeds while remaining consistent with CMB spectral distortion constraints, and examine the role of accretion history in constraining these models. We perform a suite of Markov chain Monte Carlo analyses of power spectrum enhancements (lognormal and cut-power-law) with varying local non-Gaussianity and accretion histories, satisfying the COBE/FIRAS $\mu$-distortion constraint with and without the requirement that the model produces SMBH seeds. We find that both models can seed LRDs and evade $\mu$-distortion bounds for $|f_{\rm NL}|= O(1)$ and peak amplitude $A= O(10^{-3})$, when standard Eddington-limited accretion is also incorporated. Interestingly, the results exhibit a preference for both positive and negative values of $f_{\rm NL}$. The preference for negative $f_{\rm NL}$ is strongly dependent on the accretion history, and is removed if the black holes have a small duty cycle, $f_{\rm duty}\lesssim 0.1$. These results demonstrate that the JWST high-$z$ LRD population can be seeded by primordial perturbations while remaining compatible with CMB data, and highlight the importance of next-generation CMB spectral distortion missions to test this hypothesis.
Read the full abstract

All authors Shyam Balaji, Bryce Cyr, Sarah R. Geller, David I. Kaiser, Dario L. Lorenzoni, Evan McDonough, Wenzer Qin

Astrophysics of Galaxies

Search for a Dispersion Measure Excess Toward Nearby Galaxy Halos Using Fast Radio Bursts from the Second CHIME/FRB Baseband Catalog

Fast radio bursts (FRBs) probe foreground ionized baryons via their dispersion measures (DM). Using 1748 sub-arcminute-localized sources from the Second CHIME/FRB Baseband Catalog, we conduct a statistical stacking study of the circumgalactic media (CGM) of nearby galaxies by analyzing the DMs of FRBs whose sightlines intersect foreground halos. For galaxy halos within 50 Mpc, we measure a mean DM excess of $119\pm44$ pc~cm$^{-3}$ $(p = 0.0053; \sim 2.6\sigma)$ within one virial radius. Expanding to 150 Mpc (628 intersections), we find a tentative decline in the mean DM as a function of impact parameter $(p = 0.0046; \sim 2.6\sigma)$, which is predominantly driven by typical Milky Way (MW)-like galaxies $(10 \leq \log(M_*/M_\odot)\leq 11)$. We use this DM profile to constrain parameters of the CGM electron density distribution, $n_e^\mathrm{CGM}(r) \propto r^{-3\beta}$, finding an extrapolated central electron density in the range of $(1.7-21.1) \times 10^{-3}\mathrm{\,cm}^{-3}$, with a radial power-law index of $\beta = 0.68^{+0.15}_{-0.16}$, consistent with constraints from X-ray stacking studies of MW-like galaxies. At very low impact parameters, we detect a significant DM deficit likely due to cross-matching FRBs that originate from galaxies within our galaxy sample. By deriving the DM-galaxy cross-correlation function, we show that this deficit matches theoretical predictions if 20 FRBs originate from galaxies within 150 Mpc, indicating that CHIME/FRB detects a higher fraction of low-$z$ FRBs than previously thought. We forecast that $10^4$ FRBs with similar localization errors will provide a 5-$\sigma$ detection of the CGM, and that FRBs at intermediate redshifts ($0.4\lesssim z_\mathrm{FRB}\lesssim 0.75$) are ideal for studying intervening CGMs.
Read the full abstract

All authors Affan Khadir, Ayush Pandhi, Jason Hessels, Calvin Leung, B. M. Gaensler, Ludwig M. Böss, Amanda M. Cook, Alice P. Curtin, Klaus Dolag, Naman Jain, Victoria Kaspi, Afrokk Khan, Adam E. Lanman, Kiyoshi W. Masui, Daniele Michilli, Mason Ng, Aaron B. Pearlman, Sachin Pradeep E. T., Aniket Prasad, Mawson Sammons, Paul Scholz, Kaitlyn Shin, Kendrick M. Smith, Jenny G. Sorce, Haochen Wang, Michele Woodland

Quantum Physics

The hidden symmetry operators of the asymmetric quantum Rabi model and its two-photon and two-mode variants in closed form

The asymmetric quantum Rabi model, a two-level system with splitting $\Delta$ and bias $\epsilon$ coupled with strength $g$ to a boson mode, has level crossings when $\epsilon$ is an integer multiple $N$ of the mode frequency. They are due to a hidden symmetry operator $J$, for which no general formula has been available. Here we obtain $J$ in closed form for every $N$. We find that $J$ is given by the parity of the symmetric model times the continuant (ordered tridiagonal determinant) of $N$ first-order operators, which are the Hamiltonians at the intermediate biases $N-1,N-3,\dots,1-N$ minus a common number operator. The symmetry is thus a staircase of $N$ bias steps that closes on the mirror image of the model precisely at integer bias. The proof uses induction on two commutation relations. The same relations hold in the asymmetric two-photon and two-mode models, written with the generators of $\mathfrak{su}(1,1)$, so that the same continuant gives their symmetry operators for every order, previously known only for the lowest ones. We also relate the closed formula for $J$ to the known recursion relations for its coefficients and give their terms of lowest order in $\Delta$ explicitly.
Read the full abstract

Mathematical Physics

Eigenvector Occupation Statistics and Extremes in the Rosenzweig-Porter Model

We study the occupation probabilities $p_i=|u_i|^2$ of a normalized bulk eigenvector in the GUE Rosenzweig-Porter model $A_N=B_N+\tau_NG_N$, with $\tau_N^2=N^{1-\gamma}$ and $1<\gamma<2$. Conditioning on an eigenvalue at a regular bulk energy $E$ through the normalized one-eigenpair Kac-Rice law, we derive an exact finite-$N$ projective density and reduce its asymptotic analysis to a Hubbard-Stratonovich integral over $2\times2$ Hermitian matrices, which we evaluate by the saddle-point method. The resulting occupation field is governed by a self-consistent Lorentzian profile with effective resonant size $K_N(E)\asymp N^{2-\gamma}$ and local exponential fluctuations. We prove the asymptotics of all generalized inverse participation sums and identify a transition at $q=1/2$. We also obtain a two-branch thickened amplitude-counting spectrum. At the two ends of the occupation field we prove marked Poisson point-process limits: the largest components have Gumbel fluctuations, whereas the smallest component has an exponential limit on the scale $N^{-1-\gamma}$. Finally, for every fixed $q>0$ the largest component contributes an asymptotically vanishing fraction of the inverse participation sum, showing that these fixed-$q$ statistics do not freeze. The statements are quenched, uniformly on a high-probability regularity event for the diagonal disorder, and hence also hold in the annealed setting.
Read the full abstract

General Relativity and Quantum Cosmology

Mapping of Spacetimes with Neural Networks

Gauge freedom in numerical relativity complicates the interpretation of simulation coordinates and the comparison of different descriptions of the same spacetime. We present a data-driven neural network method for learning coordinate maps from numerical metric data to a prescribed target gauge. The network is trained by matching the transformed source metric to the target metric, without assuming a functional form for the coordinate transformation. We first validate the method on analytical Schwarzschild and Kerr geometries, recovering known transformations up to the isometries of the target spacetime. We then apply it to numerical black hole evolutions in moving-puncture coordinates. For a non-spinning black hole, the learned map relates the time-dependent numerical metric to a stationary Schwarzschild metric in ingoing Eddington-Finkelstein coordinates. For a spinning puncture, the method provides an approximate identification with a Kerr background despite departures from Kerr in the numerical spacetime. We use this identification to construct curvature perturbations and evolve them with the Teukolsky equation, obtaining waveforms that closely match those extracted from the full nonlinear simulation. These results demonstrate that learned coordinate maps can relate numerical spacetimes to analytical backgrounds, with potential applications to cross-code validation and black hole perturbation theory.
Read the full abstract

All authors Estuti Shukla, David Radice, Rossella Gamba, Sebastiano Bernuzzi

Computation and Language

WorkflowArena: Benchmarking Executable Workflow Orchestration in LLM-Based Multi-Agent Systems

Large language model agents are increasingly deployed as multi-agent systems rather than as a standalone model operating within a single context. Recent agent harnesses allow the model to instantiate worker agents and to author the executable workflow that coordinates them, shifting the design of system structure from developers to the model at run time. The quality of this organization therefore constitutes a distinct capability, yet existing benchmarks either measure end-to-end task completion or evaluate coordination under structures specified in advance. We formalize this capability as workflow orchestration and introduce WorkflowArena, a benchmark of 145 long-horizon tasks spanning 8 professional domains, from software engineering and formal mathematics to legal analysis and 3D production. Each task specifies an initial workspace and an integrated deliverable composed of interdependent components, which is assessed by a hidden verifier. Agents determine autonomously whether and how to delegate, and all delegation is recorded, enabling joint evaluation of deliverable quality and the organization of the underlying work. Experiments with nine models show that no single configuration performs best across all domains and that systems exhibit distinct delegation strategies: at high effort, Claude models in Claude Code invoke workflows less frequently but with larger worker groups, whereas GPT models in Codex delegate more frequently in smaller units. When the executing model is held fixed on a challenging subset of tasks, workflows authored by Claude Opus 5.5 substantially improve weaker executors, indicating that workflow organization is itself a measurable determinant of task performance.
Read the full abstract

All authors Zheng Jiang, Fang Niu, Bo Peng, Houde Qian, Wenyi Li, Deyao Hong, Mingju Gao, Yizhe Chi, Youjie Zheng, Kaisen Yang, Calvin Xiao, Lifeng Sun, Huan-ang Gao

Computer Vision

AI-based characterization of lymphoid structures and the tumor microenvironment reveals lower mature tertiary lymphoid structures, lymph-node follicles, and plasma cells after neoadjuvant therapy in lung squamous cell carcinoma

Mature, germinal centre-bearing tertiary lymphoid structures (TLS) and mature lymph node follicles (secondary lymphoid structures, SLS) sustain the plasma cell response that shapes outcome and immunotherapy benefit in lung cancer. How this loco-regional axis differs between neoadjuvant-treated and therapy-naive tumors is poorly defined. We analyzed H&E whole-slide images of resected tumors and paired proximal and distal lymph nodes from matched neoadjuvant-treated (neoLUSC, n=20) and therapy-naive (naivLUSC, n=20) lung squamous cell carcinoma (LUSC) patients. Aignostics Atlas H&E Tissue Analysis (HETA) models quantified cell densities per tissue compartment and cell-cell co-localization; a dedicated deep-learning model validated against pathologist annotations detected TLS and SLS and graded their maturity. Prognostic value was tested in an independent cohort (n=458; 46 treated). In neoLUSC tumors, stromal plasma cell density was 72% (p=0.004) and stromal lymphocyte density 48% (p=0.015) lower than in naivLUSC; macrophages and granulocytes did not differ significantly. Total and immature TLS did not differ, whereas mature TLS in the tumor (p=0.034) and mature SLS in the tumor-distal lymph node (p=0.003) were lower in neoLUSC patients. Peri-TLS plasma cell density correlated with tumor lymphocyte density in naivLUSC (rho=0.54, p=0.017) but not in neoLUSC tumors (rho=0.07). In the validation cohort, stromal plasma cell density (HR 0.69, p=0.011) and lymphocyte-plasma cell co-localization (HR 0.66, p=0.004) were associated with longer survival, and neoadjuvant status with shorter survival (HR 1.50, p=0.026). Neoadjuvant therapy is thus associated with fewer mature lymphoid structures and weaker coupling of the humoral compartment to the tumor rather than uniformly lower immune readouts. The stromal plasma cell/co-localization axis is a candidate prognostic biomarker.
Read the full abstract

All authors Arne Schmidt, Justin Nieder, Cornelius Boehm, Julika Ribbat-Idel, Marius Teodorescu, Marie-Lisa Eich, Simon Schallenberg, Gabriel Dernbach, Katja Lingelbach, Miriam Haegele, Timo Milbich, Adam Narai, Maria Stoelben, Hans Pinckaers, Lukas Ruff, Kai Standvoss, Christian Watermann, Christiane Kuempers, Sven Perner, Verena Sailer, Jasmin Tiebach, Torsten Goldmann, Sebastian Marwitz, Johannes Braegelmann, Frederick Klauschen, Marija Pezer, Rosemarie Krupar

Cryptography and Security

TUTTI: Parallel CKKS Bootstrapping in under 3 ms on CPU

This paper introduces TUTTI, a new CKKS bootstrapping procedure that outperforms previous low-latency solutions in multi-threaded latency for any number of messages and, most notably, has a lower latency than TFHE when bootstrapping a single one. Our bootstrapping builds upon SPRU (CK, TCHES 2027), improving it with new methods for exploiting parallelism and increasing the number of messages, as well as with a refined security analysis, of independent interest, of the block-binary secret keys of LMSS (AsiaCCS 2023). TUTTI bootstraps 4096 real-number messages in 46 to 83.4 ms on 192 threads to restore 1 to 9 multiplicative levels, outperforming SHIP (CHKS, Eurocrypt 2025) by a factor of up to 3 on the same machine. Unlike SHIP, TUTTI gains immediately from bootstrapping fewer messages, with a 256-message bootstrapping taking just 15 to 34 ms on 32 threads. For a single message, TUTTI takes 4.6 ms on 32 threads, whereas TFHE-rs needs 7.3 ms for a 4-bit bootstrapping on the same machine and thread count. We show that this advantage persists across different architectures: on an Apple M5 Pro laptop (ARM), TUTTI takes 2.4 ms on 14 threads, against 4.5 ms for TFHE-rs. Differences between CKKS and lightweight schemes such as TFHE remain on several fronts, from functionality to key sizes and security assumptions, but our work shows that multi-threaded latency is no longer one of them, closing a long-standing gap in the practical FHE literature.
Read the full abstract

Probability

Distributional Limits of Empirical Codegree Processes in Sampled Networks

We study fluctuations of empirical distributions of normalized codegrees under independent vertex sampling from deterministic or random weighted parent networks. We establish a quantitative multivariate normal approximation for sampled biclique counts. Under additional graph-limit and sparsity assumptions, we establish convergence of the finite-dimensional distributions of the empirical codegree process indexed by smooth test functions. We further establish functional convergence on Hölder balls, with a smoothness threshold depending on network sparsity and vertex sampling probabilities. We verify our assumptions for classes of graphon-generated random graphs and give counterexamples showing that the smoothness conditions cannot in general be weakened.
Read the full abstract

Computer Vision

What kills $v$-prediction? A Patch-wise PCA Perspective on Pixel-Space Flow Matching

Latent Diffusion Models dominate generative modeling but face two-stage training complexity and the inherent reconstruction limits of pre-trained autoencoders. Pixel-space Flow Matching offers a simplified end-to-end alternative yet often encounters training difficulties when using the standard $v$-prediction objective. Recent research attributes this difficulty to the off-manifold nature of noised targets, motivating a shift toward $x$-prediction. In this paper, we investigate this explanation through patch-wise Principal Component Analysis (PCA) and identify a bandwidth bottleneck. The velocity in minor components is largely determined by the raw noisy input, requiring deep backbones to preserve information that competes with principal-component modeling. Through controlled experiments and representation analysis, we further examine why $x$-prediction remains effective: its parameterization provides a raw-input bypass, while its learned input representation can spontaneously concentrate on principal directions. Under low-rank observations, our analysis of the induced flow characterizes $x$-prediction as an implicit conditional-mean decoder, reconstructing discarded components from the retained information. These observations motivate the Raw-injected Transformer (RiT), which combines a lightweight raw-conditioned decoder with an output-side shortcut to separate semantic modeling from raw-state propagation. Our method restores effective $v$-prediction and achieves competitive performance on ImageNet $256\times256$, providing a perspective on the interaction between prediction objectives and architectural bandwidth.
Read the full abstract

Systems and Control

Distributed Global Pose Synchronization for Heterogeneous Multiple Rigid Body Systems on SE(3)

This paper addresses the problem of global pose synchronization for heterogeneous multiple rigid-body systems evolving on the matrix Lie group $\SE(3)$. We first construct a new potential function on $\SE(3) \times \mathbb{R}$ with a generalized non-diagonal weighting matrix, together with a set of switching variables governed by continuous-discrete hybrid dynamics. Based on this new potential function and these switching variables, we propose a distributed hybrid feedback control scheme that uses relative pose on $\SE(3)$ and group velocity measurements, endowed with global asymptotic stability guarantees. We then further propose a distributed velocity-free hybrid feedback control scheme that uses relative and absolute pose measurements, without requiring group velocity measurements, while preserving global asymptotic stability. Numerical simulations are presented to demonstrate the performance of the proposed distributed hybrid control schemes.
Read the full abstract

All authors Fengyu Lin, Miaomiao Wang, Housheng Su, Zhigang Zeng, Abdelhamid Tayebi

Mesoscale and Nanoscale Physics

Enhancing valley splitting in Si/SiGe Wiggle Well quantum dots through integrated stressor-gate architectures

Engineering large and reproducible valley splittings is essential for the scalable operation of silicon spin qubits. In Si/SiGe quantum dot devices, valley splitting arises predominantly from atomic-scale disorder and is therefore inherently random, motivating approaches that introduce a large deterministic contribution. Strain-assisted Wiggle Wells provide a route to deterministic valley splitting by combining an oscillatory Ge concentration in the Si well with shear strain; however, generating the required strain in a manner compatible with large-scale architectures remains a challenge. Here, we propose patterned stressed thin films as a scalable approach for generating shear strain in the quantum well of Si/SiGe quantum dot devices. Finite-element simulations show that compressively stressed TiN films can generate shear strains exceeding 0.1% at the quantum well, sufficient for deterministic valley splittings of hundreds of $\mu \text{eV}$ in strain-assisted Wiggle Wells. We show that this strain can extend across multiple quantum dots and is maintained in realistic gate geometries, providing a fabrication-compatible approach for scaling deterministic valley splitting to quantum dot arrays.
Read the full abstract

All authors E. S. Joseph, Talise Oh, D. E. Savage, D. S. Stone, DeAnna Campbell, Benjamin D. Woods, Mark Friesen, M. A. Eriksson

Earth and Planetary Astrophysics

The Tierras Observatory: Ultra-Precise Photometry for Characterizing Exoplanets and Their Host Stars

The Tierras Observatory is a fully robotic 1.3-m telescope at the Fred Lawrence Whipple Observatory in southern Arizona. Tierras achieves sub-part-per-thousand (ppt) photometric precision on two timescales: intra-night, to detect and characterize transits of planets around nearby mid-to-late M dwarfs, and night-to-night across an observing season, to characterize the long-term variability of their host stars. We describe the facility and instrument design, our photometric pipeline, and our on-sky performance. We present the first ground-based light curves with sub-ppt precision over timescales of months. This precision enables a range of exoplanet-related science, including the measurement of host star rotation periods, the detection of long-period transits, and the discovery of additional terrestrial planets in known M dwarf systems.
Read the full abstract

All authors Juliana García-Mejía, Patrick Tamburo, David Charbonneau, Daniel Fabricant, Jonathan M. Irwin, Robert Fata, Joseph M. Zajac, Peter E. Doherty, Brian McLeod, Pascal Fortin, Michael L. Calkins, Gilbert A. Esquerdo

Logic

Adding unbounded and not bounded sets to a singular cardinal

This paper solves a problem (SP) attributed to Solovay in the 1960s, which is to add an unbounded sequence to a singular cardinal $\kappa$ without adding bounded subsets to $\kappa$, while preserving $\kappa^+$. The solution applies to any cofinality of the singular cardinal, with the assumption that $\kappa$ is a strong limit needed for the case of uncountable cofinality. Unsuccessful attempts to solve (SP), mostly for the countable cofinality, have appeared in published papers by various authors.
Read the full abstract

2611.12848 · 9 Nov · After a helpful comment by Mohamammad Golshani, we were able to change and simplify the proof of Lemma 3.5, improving the theorem. The previous version of this lemma had a mistake which made it appear as if we needed to use ultrafilters, where only filters are needed (and the singleton filter suffices)

Algebraic Topology

Discrete Configuration Spaces of Simplex Skeleta: a Middle-Dimensional Manifold Phenomenon

For a simple graph $\Gamma$ without loops and an integer $n\geq2$, Abrams showed in his Ph.D. thesis that the discrete configuration space $DConf(\Gamma,n)$ is homeomorphic to a closed $n$-manifold if and only if $n=2$ and $\Gamma$ is $K_5$ or $K_{3,3}$, the two minimal obstructions to graph planarity. We show that this manifold property persists for the infinite family of discrete configuration spaces obtained by replacing the Kuratowski graphs with the $d$-skeleton of the simplex of dimension $2d+2$, thus recovering Abrams' $K_5$ case when $d=1$. Our manifold recognition argument relies on a recursive application of Whitehead's theorem that collapsible PL manifolds are PL balls, combined with Forman's discrete Morse theory. The resulting construction produces an infinite family of combinatorial spheres that contains the Legendre polytope boundaries and is closed under taking vertex links.
Read the full abstract

Astrophysics of Galaxies

Forward modeling linear variable filter data: High resolution PAH and Paschen-$\alpha$ maps from SPHEREx

Polycyclic aromatic hydrocarbon (PAH) emission is a powerful tool for the study of the interstellar medium (ISM) in both Galactic and extragalactic contexts. It can reveal ionization state, radiation field, metallicity, composition, and ISM structure as sculpted by processes like turbulence and stellar feedback. Here we use NASA's SPHEREx to produce clean, high-resolution PAH maps at spectral resolution R $\sim$ 35 (compared to WISE at R $\sim$ 2). Prior treatments of the SPHEREx linear variable filter data seek to convert them to more conventional band-averaged maps that can leave striping artifacts associated with spatially variable wavelength sampling, especially at emission lines. In contrast, we deliver a scalable forward-modeling method capable of generating high-quality all-sky SPHEREx emission-line maps at 8.5 arcsec FWHM angular resolution (on 6.4 arcsec pixels), while naturally producing maps without visible striping. We reformat the data into a pixel-level database, mask stellar contamination, and infill missing pixels with a conditional, data-driven, anisotropic Gaussian process. We apply our method to produce 3.3 $\mu$m and 3.4 $\mu$m PAH and Paschen-$\alpha$ emission maps of a set of fields spanning a range of Galactic and extragalactic conditions including the Rosette nebula, the Corona Australis molecular cloud, M33, and M31. We demonstrate the quality and synergies of our SPHEREx 3.3 $\mu$m PAH maps by comparisons with other available gas and dust tracer maps and achieve sensitivities up to $\sigma(I_{3.3}) = 2.7 \times 10^{-8}$ erg s$^{-1}$ cm$^{-2}$ sr$^{-1}$. We estimate that a two-year extension of the SPHEREx survey, with modest modifications to the scanning strategy, would match or exceed the PAH sensitivity of WISE over the whole sky.
Read the full abstract

All authors Andrew K. Saydjari, Connor Hainje, Benjamin Dodge, Philipp Frank, David W. Hogg, Susan E. Clark

High Energy Astrophysical Phenomena

AGN Feedback Turbulence in Cool-core Clusters: Evidence from XRISM

The XRISM X-ray Observatory has revealed evidence for a "double-cascade" scenario in the Perseus cluster, in which active galactic nucleus (AGN) feedback and mergers drive turbulence over distinct radial ranges and spatial scales. Whether this gas kinematic feature is common among cool-core clusters remains unclear. Here, we revisit XRISM observations of A1795 and A2029 to investigate the signatures of AGN feedback and their contribution to the gas velocity fields. In A1795, AGN feedback dominates the velocity dispersion in the core, similar to Perseus, and may provide turbulent heating at a rate comparable to radiative cooling. In contrast, we obtain only an upper limit on the AGN feedback contribution in A2029, where merger-driven motions dominate. Tailored simulations indicate that an off-axis minor merger $\simeq1.6$ Gyr ago displaced the central low-entropy gas in A2029, potentially suppressing AGN activity. This gas is now returning to the center, possibly facilitating the re-establishment of AGN feedback. We propose an evolutionary view of the double-cascade scenario, in which merger-driven motions dominate during and shortly after significant mergers, while feedback-driven turbulence re-emerges in cluster cores as systems relax. This framework provides a basis for interpreting XRISM kinematic measurements of (weak) cool-core clusters. Our findings highlight the importance of spatially resolved gas kinematics and detailed modeling for identifying double-cascade features and disentangling the contributions of AGN feedback and mergers.
Read the full abstract

All authors Congyao Zhang, Eugene Churazov, Irina Zhuravleva, Hannah McCall, Annie Heinrich, William R. Forman, Ildar Khabibullin, Michal Zajaček, Norbert Werner

Representation Theory

On $\mathrm{HH}^1$ of blocks of finite simple group algebras

Let $G$ be a sporadic simple group, $k$ a field of positive characteristic and assume $B$ is a block of $kG$ with non-trivial defect. In this paper we prove the first Hochschild cohomology group $\mathrm{HH}^1(B)$ is non-vanishing. Furthermore, we prove that for every wild block with abelian defect of a sporadic simple group, we have that $\mathrm{HH}^1(B)$ is a solvable Lie algebra of derived length two, with non-zero maximal toral rank.
Read the full abstract

Distributed Computing

Contention Resolution with a Fast-but-Fickle Channel

The contention-resolution problem studies how $n$ devices share a communication channel without centralized coordination. Classical models typically assume a single continuously available channel and focus on minimizing total completion time. However, modern wireless systems increasingly rely on shared spectrum, where access may be abruptly revoked by incumbent users. Motivated by this setting, we study contention resolution with two channels: a reliable slow channel and a faster channel that can be intermittently disabled by unpredictable eviction periods. Moving a batch of packets from the slow channel to the fast channel incurs a unit cost, while evictions from the fast channel impose a larger per-packet penalty. Our goal is to achieve asymptotically optimal latency while minimizing switching overhead relative to an optimal offline algorithm that knows the eviction schedule. We present Randomized Adaptive Channel Exploitation (RACE), a randomized algorithm that exploits the fast channel when available while maintaining progress on the slow channel during disruptions. When the fast channel has rate $\rho \in \bigl[\Omega(\log n), O(\sqrt n)\bigr]$ and a sufficiently large constant-factor resource augmentation, RACE achieves an $O(1)$ competitive ratio for expected latency. RACE's expected switching-cost competitive ratio relative to any optimal-latency offline algorithm OPT is $O(\rho+\sqrt{n/m})$, where $m \geq 1$ is the number of fast-channel availability intervals used by OPT. We also prove a worst-case $\Omega(\sqrt n)$ switching-cost competitive ratio for latency-competitive randomized algorithms, and a separate $\Omega(\rho)$ switching-cost competitive-ratio lower bound. These lower bounds hold even with fixed constant-factor augmentation on both channels.
Read the full abstract

All authors Umesh Biswas, Vini Chaudhary, Shafqat Hasan, Maxwell Young

Differential Geometry

Delaunay Asymptotics for the Singular Escobar-Yamabe Equation

In dimensions $n\geq3$, we prove that near an isolated boundary singularity a positive solution of the Escobar–Yamabe equation for a smooth metric approaches a fixed periodic Delaunay solution or the cylindrical solution if its trace lies between positive multiples of $|\overline x|^{-(n-2)/2}$, where $\overline x$ is the tangential Fermi coordinate. The relative error decays at a positive power rate, with weighted estimates through two derivatives in Fermi coordinates. The necksize is unique, and the translation parameter is unique modulo the least positive period in the periodic case. The Hamiltonian smoothly parametrizes the periodic family modulo translation. We classify Jacobi fields of polynomial growth at both ends without assuming symmetry. Translation and energy differentiation generate these fields along periodic solutions, and two oscillatory fields generate them along the cylindrical solution. We also exhibit a first-order obstruction to improving these asymptotics by boundary-preserving conformal deformations of the flat models.
Read the full abstract

Earth and Planetary Astrophysics

Joint Gaia-Hipparcos astrometry, RV, and Imaging posteriors from the G23H Hipparcos sample

We present uniform posteriors indicating the presence or absence of stellar and planetary companions over a sample of $10^5$ stars in Hipparcos and Gaia. We use the "G23H" catalog to fit a model that incorporates absolute astrometry constraints from Hipparcos, Gaia DR2, and Gaia DR3, as well as Gaia DR3 photometry, excess RV variability, and includes information from the presence or absence of common-proper motion companions. The resulting posteriors can be used directly to detect new companions, to target RV and imaging follow-up observations, and for downstream demographics studies. We highlight example detections of known and new candidate companions and the companion detection rate across the HR diagram. The modelling approach we present is directly applicable to Gaia DR4 epoch astrometry and RVs. All posteriors are provided as a machine-readable catalog and user-accessible website http://exopo.st/.
Read the full abstract

All authors William Thompson, Dori Blakely, Alexandre Bouchard-Côté, Trevor Campbell, Jerry Xuan, Jingwen Zhang, Kyle Franson, Jessica Speedie, Clarissa Do Ó, Doug Johnstone, Eric Nielsen, Brendan Bowler, Simon Blouin, Jean-Baptiste Ruffio, Ryan Cloutier, Andre Fogal, Kaitlyn Hessel, Christian Marois, Alexandra Rochon, William Roberson

Materials Science

High Pressure Material Phase Identification with Gaussian Kernel Neural Networks

Phase identification is a computational task involving the classification of material structure using the positions of atoms. Using data from Molecular Dynamics (MD) simulations, our task is to identify the material phase of a set of atoms, based on how the simulations predict their positioning relative to each other. By leveraging existing research on mechanistic neural network theory, particularly Neural Tangent Kernels (NTK), we develop a new and efficient method for classifying simulated material samples. Our method involves defining a set of "atom detectors" that are given positions around the origin in cartesian coordinates. The sum of Gaussians centered at each atom is calculated at each atom detector, and is used as the input to a neural network. We present theoretical and experimental evidence to show that this method is highly effective and efficient at the task. Our method obtains better or similar accuracy scores to commonly used methods while having a major computational speedup.
Read the full abstract

General Relativity and Quantum Cosmology

Beyond $\Lambda$CDM: Probing Structure Growth and Observational Viability of Generalized ${f\,(Q , \mathcal{L}_{m})}$ Gravity

$f(Q, \mathcal{L}_m)$ gravity is a well-known extension of $f(Q)$ gravity, obtained through non-minimal or additive minimal coupling in the action's Lagrangian density. Background-level cosmography within this framework, along with comparisons to $\Lambda$CDM, has been the focus of prior studies. The present work considers a generalized power-law form of the additive minimal coupling, $f(Q,\mathcal{L}_m) = -Q^\mu + \alpha \mathcal{L}_m^{\,\nu} + \beta$, from which the Hubble function $H(z)$ is derived through a novel parametrization. When $\nu \neq 1$, the matter continuity equation is rendered non-conserved, a feature that can be interpreted as an effective particle-creation mechanism. Constraints on the model parameters are obtained from Observational Hubble Data, DESI DR2 BAO measurements, and three SNe Ia compilations (Pantheon+, Union3, DES-SN5YR), allowing their cosmographic effects to be examined. Beyond the background level, the perturbed field, Euler, and continuity equations are derived under non-conservation, and their impact on structure growth is studied through a joint analysis incorporating the $f\sigma_8$ probe. Bayesian evidence, computed relative to $\Lambda$CDM, shows that the model is favoured by background data in several cases, with moderate support for one realisation, whereas the inclusion of $f\sigma_8$ drives $\nu\to1$ and decisively disfavours the additional freedom, highlighting the importance of growth data in testing such extensions.
Read the full abstract

General Relativity and Quantum Cosmology

Scalar quasinormal modes of the exact spinning Morris–Thorne wormhole

We study the quasinormal modes (QNMs) of a minimally coupled massless scalar field on the exact spinning Morris-Thorne wormhole. The problem is posed on the complete two-ended geometry, whose asymptotic rest frames rotate relative to each other. After extracting the axis behaviour and the outgoing phases at both infinities, we compactify the radial coordinate and construct a quadratic eigenvalue problem by two-dimensional Chebyshev collocation. The analytic background yields exact constraints on the spectrum. At fixed azimuthal number $m$, the frequencies $\widehat{\omega}=r_0\omega$ are mirror symmetric about $\operatorname{Re}\widehat{\omega}=3\pi jm/2$, where $r_0$ is the throat radius and $j=J/r_0^2$ is the dimensionless angular momentum. Each multipole family depends on $m$ and $j$ only through $mj$ and three angular Chebyshev polynomials represent exactly the multipoles $L=|m|,|m|+1,|m|+2$. A conserved flux confines nonthreshold real quasinormal frequencies to the superradiant window $0<\omega/(m\Omega_-)<1$, where $\Omega_-=3\pi J/r_0^3$. For the sampled sectors $m=1,2,3$ and spins up to $j=19$, every converged frequency is damped and lies outside this window. Independent integration verifies the low-lying modes, while Kerr spectra provide a separate benchmark for the two-dimensional solver. A limiting radial equation explains why the computed mode frequencies approach the upper edge of the superradiant window with frequency offsets proportional to $(mj)^{-1/2}$ at large $mj$. A complete stability assessment requires further analysis of the unresolved scalar spectrum and of perturbations of the coupled metric-matter system.
Read the full abstract

Statistical Methodology

Change-point analysis of textual data: a flexible approach

We propose a CUSUM-based change-point procedure for word-frequency vectors that accounts for the variation in the amount of observed textual information over time. Under a multinomial model, we derive the asymptotic null distribution of the proposed statistic, allowing for varying document numbers and lengths. With trimming based on cumulative word counts, the limiting null distribution does not depend on the temporal distribution of textual information. We also consider parametric and nonparametric bootstrap procedures for finite-sample calibration. A simulation study examines their performance under both multinomial sampling and Dirichlet mixtures of multinomial distributions. The results demonstrate the advantages of the proposed approach under increasing document intensity and show that resampling entire documents provides more reliable calibration in the presence of document-level heterogeneity. Finally, we apply the procedure, combined with binary segmentation, to 68,542 abstracts from the arXiv category stat.ML spanning 2007-2025, identifying multiple changes in word-usage patterns.
Read the full abstract

Quantum Algebra

Property F implies weak integrality for non-degenerate pivotal braided fusion categories

We prove that every non-degenerate pivotal braided fusion category with property F is pseudo-unitary and weakly integral. In particular, every modular category with property F is weakly integral. Neither sphericality nor pseudo-unitarity is assumed. Passing to the Drinfeld centre extends the result to degenerate categories whose dimensions are positive on the intersection of the Müger centre with the adjoint subcategory; these include every braided fusion category with a pseudo-unitary Galois conjugate. Physically, the result connects the arithmetic of topological orders to the computational power of anyon braiding: non-integer squared total quantum dimension guarantees infinite braid group images, a necessary condition for universal quantum computation.
Read the full abstract

All authors Justin Asher, Jonah Berean-Dutcher, Clay Cordova, Julio Parra-Martinez, Richard Tuddenham

Machine Learning

Mitigating Agentic Deception via Evaluation-Side Partially Observable Games

As large language models become more capable agents, they can exhibit deceptive behaviors, cheating human for higher rewards while appearing helpful. We attribute this failure to evaluation-side partial observability, where models are evaluated mainly from final outcome, without checking the faithfulness of underlying reasoning, tool use, and environmental effects. Prior mitigations address this partial observability by predefined evaluator latent belief space or one-shot monitoring, which do not extend to free-form language space and vulnerable to stronger deceptions that evade simple monitoring. In this work, we formulate deception mitigation as an Evaluation-Side Partially Observable Extensive Form Game (e-PoG), a max-min framework in which an agent maximizes reward, a monitor minimizes the agent's payoff by exposing deception from the full interaction trajectory, and a fixed verifier validates the monitor's claims and converts them into penalties. We define the solution concept as a Subgame-Perfect Oversight Equilibrium (SPOE) between agent and monitor, and prove its existence in finite games. With a perfect verifier, every SPOE is faithful; with an imperfect verifier, the guarantee persists if verifier errors do not reverse the ranking of monitor actions. To achieve faithful SPOE, we derive a two-timescale policy gradient algorithm and enhance the pipeline empirically with NLI verification and deterministic checks. Experiments with Gemma3-4B-IT and Qwen3-4B/8B on Agentic Role-Playing, SearchQA, and WebShop show that e-PoG mitigates multiple emergent deception modes, outperforming the strongest baselines by 26.0% in deception mitigation and by up to 39.8% in task success. These gains persist under weak-to-strong supervision. Code is available at https://github.com/Toyoid/deceptive-agent.
Read the full abstract

All authors Hanxiao Li, Shutong Liu, Yongjia Ma, Jiaming Ji, Juntao Dai, Shuning Zhang, Kejian Shi, Kai Chen, Pu Feng, Yuqing Ma, Qi Dou, Yaodong Yang, Xianglong Liu, Simin Li

Human-Computer Interaction

Programming Prescience: Issues in Human Construction of Forecasting Bots

AI models' growing capacity for complex reasoning has made them attractive tools for forecasting, or the systematic prediction of the future based on past and present data. However, should AI forecasting systems be adopted, human builders will shoulder the work of building and evaluating them. In this paper, we explore what this work looks like with a qualitative interview study with 15 builders of forecasting bots, sampled from among successful participants in a recent AI forecasting competition. Our analysis surfaces the challenges around (1) tuning a bot to acquire factual research, (2) implementing coherent reasoning, and (3) understanding how well one's bot actually does. During their work, builders frequently introduced complexity that did not pay off, and tested in ways that were opportunistic and piecemeal. Our detailed account of their challenges helps clarify what problems should be targeted in upcoming innovations in process, tools, and human-AI teaming for forecasting.
Read the full abstract