High Energy Physics (Theory)
Deterministic Hypergraph SYK: Melonic Dominance and Maximal Chaos Without Disorder Averaging
Shiroman Prakash
We show that the leading melonic structure of the Sachdev-Ye-Kitaev (SYK) model can be realized by a single sparse Majorana
Hamiltonian, without disorder averaging. In our deterministic hypergraph SYK model (hSYK), $N$ Majorana
fermions interact through $O(dN)$ terms with uniform couplings, arranged on a fixed $q$-uniform, $d$-regular hypergraph. If the hypergraph is linear, has few short cycles, and is a spectral expander, then at large $d$ the two-point function obeys the melonic Schwinger-Dyson equation site by site. The leading thermodynamics are therefore those of SYK, and at leading order in $1/d$ every bilocal fluctuation other than the Schwarzian is gapped. The four-point ladder splits into scrambling modes labeled by
eigenvalues of the site adjacency operator. At strong coupling, the uniform mode is maximally chaotic, and a non-uniform mode grows only if its
eigenvalue exceeds $1/(q-1)$. Exact diagonalization of single instances with $N\le24$ agrees with the melonic two-point function and thermodynamics more closely than dense SYK of the same size, shows Wigner-Dyson level statistics, and reproduces the predicted mode structure of out-of-time-order correlators. Numerically, random linear hypergraphs typically satisfy these conditions, and for $d>q$ also avoid emergent
symmetries. The model is a natural candidate for
quantum simulation.
Read the full abstract ▾
Quantum Physics
Emergent classicality and wavefunction branching in an isolated quantum many-body system
Saúl Pilatowsky-Cameo, Jordan Cotler, Daniel Ranard et al. (+1 more)
Decoherence in
quantum systems is conventionally modeled as the effect of interactions with an external environment. However, such a prescription excludes isolated many-body systems, which are also expected to display classical behavior at macroscopic scales. In isolated systems, decoherence must emerge internally from microscopic degrees of freedom that are invisible to the macroscopic description. Here we explicitly show that classicality can emerge in such a fashion. We consider a weakly disordered, $3$-local chaotic kicked top of $N$
qubits, where the collective
spin sector serves as the macroscopic description, while the microscopic permutation sector acts as an internal bath, decohering the collective
spin sector. Starting from closed
unitary dynamics, we derive and numerically confirm an effective Lindblad equation for the collective
spin variables. In the thermodynamic limit these reduced dynamics converge to a classical chaotic Fokker–Planck equation with vanishingly small
diffusion on the spherical phase space, producing a
quantum-classical correspondence beyond the Ehrenfest time. The chaotic dynamics evolve the pure many-body wavefunction into continuously branching components associated with distinct classical trajectories. These branches acquire nearly orthogonal microscopic records in the permutation sector, preventing
quantum interferences and ensuring the corresponding histories remain consistent.
Read the full abstract ▾
All authors Saúl Pilatowsky-Cameo, Jordan Cotler, Daniel Ranard, C. Jess Riedel
Astrophysics of Galaxies
A High-Mass Size Deficit for Void Galaxies in the Complete SDSS DR7
Olivia Curtis, Bryanne McDonough and Tereasa G. Brainerd
We characterize how environment shapes the sizes, luminosity functions, and mass functions of
galaxies in large-scale underdensities by comparing the NSA catalog of the Sloan Digital Sky Survey Data Release 7 with an identically selected TNG300 sample, both built with the void finding
algorithm VoidFinder, and by assigning a local average underdensity contrast to each
galaxy via a spherical top-hat smoothed density field approximation. Void
galaxies have fainter characteristic magnitudes than their non-void counterparts, and, at fixed
stellar mass,
redshift, and color, NSA void
galaxies with $M_*>10^{11}h^{-1}M_\odot$ are about $11\pm3\%$ more compact than
galaxies in the field, a deficit the TNG300 simulation reproduces. The
galaxies that are responsible for this effect are almost all central and predominantly early type, i.e., the systems that are most likely to grow their outer envelopes via late-time mergers. These trends hold across two
redshift bins to $z\leq0.114$. Crucially, this deficit is not a fixed property of the void sample but a steep function of the underdensity contrast, deepening toward the emptiest interiors and washing out at a typical density, implying that strict density control and sample selection are imperative when conducting environmental dependency studies like these. The same environmental dependence appears in the
stellar mass function, which shifts to lower masses in the deepest voids, and in the close-pair (i.e., the ongoing merger) fraction, which falls toward the emptiest regions. Together, these point to a late-time, merger-driven growth of massive
galaxies that is suppressed in voids, leaving their most massive members structurally distinct today.
Read the full abstract ▾
High Energy Physics (Theory)
Cosmological Correlators from Scattering Amplitudes: A Review
Soner Albayrak and Savan Kharel
The
primordial statistics underlying the CMB and large-scale structure are encoded in late-time cosmological correlation functions (constructed from the coefficients of the wavefunction of the universe). They are the closest analogue cosmology has to an $S$-
matrix, yet the standard $S$-
matrix toolkit does not directly apply. This review explains how the modern scattering-amplitudes program (spinor helicity, on-shell recursion, the double copy, positive geometry, and generalized unitarity) has nevertheless been extended to cosmology and, through a single analytic continuation, to Euclidean anti-de Sitter space, where the same objects appear as
holographic boundary correlators and are often cleanest to compute. The organizing framework is momentum space on the late-time boundary, where the
primordial statistics are simplest and where the total energy of the external legs, the sum of their momentum magnitudes, controls the analytic structure. At the total-energy singularity, the flat-space scattering amplitude emerges. We develop this toolkit starting from the wave equation and use it to compute tree-level scalar, gauge-field, and graviton correlators, exposing relations across
spins and dimensions. We then show how bulk integrals can be bypassed using cosmological polytopes, on-shell recursion, and curved-space versions of the double copy, before extending the discussion to loops, soft limits, transition amplitudes, and (A)dS cutting rules. Finally, we connect these structures to inflationary observables and to three allied programs: the cosmological optical
theorem, the cosmological bootstrap, and the cosmological
collider. We conclude with a map of the current frontier and a set of open and interesting problems.
Read the full abstract ▾
Quantum Physics
Theory of post-selected entanglement transitions in monitored bosons
Ilia Komissarov, Emanuele G. Dalla Torre and Ahana Chakraborty
Entanglement phase transitions driven by
quantum measurements have emerged as a central paradigm in open
quantum many-body physics. Such phase transitions are well established for systems with finite local
Hilbert-space dimensions, such as
qubits and
fermions, while their realization in bosonic systems with unbounded local occupation numbers remains poorly understood. Even in the absence of interactions, number states of
bosons are intrinsically non-Gaussian, preventing the use of standard correlation-
matrix approaches. To address this problem, we develop a replica-free Keldysh field-theoretic framework that expresses the Renyi
entropy of bosonic systems initialized in on-site Fock states in terms of permanents of
matrices constructed from single-particle Green's functions. Applying this framework to a continuously monitored one-dimensional cross-stitch
lattice conditioned on the no-click trajectory, we uncover a transition from volume-law to logarithmic
entanglement scaling. We show that the transition is controlled by a restructuring of the non-Hermitian spectrum that changes the number of long-lived modes from extensive to finite. In the strongly monitored regime,
bosons dynamically condense into a microscopic number of slowest-decaying modes, producing logarithmic
entanglement scaling, whereas an extensive
manifold of long-lived modes at weak monitoring gives rise to volume-law
entanglement. Our results establish a distinct mechanism for measurement-induced
entanglement transitions in free bosonic systems and provide a computationally efficient diagnostic of the measurement-induced bosonic condensation.
Read the full abstract ▾
Quantum Physics
Achieving the limits of automorphism gates
Jin Ming Koh, Shayan Majidy, Aranya Chakraborty et al. (+3 more)
Universal fault-tolerant
quantum computing combines versatile but expensive operations with specialized but cheap ones. Its efficiency depends on how much computation can be pushed onto the cheap operations and on the size of the code needed to do so.
Automorphism gates provide such cheap operations using only physical single-
qubit Clifford gates and
qubit permutations. Yet no general theory characterizes their maximum logical power or the minimum code size needed to attain it. We develop such a theory. For stabilizer codes encoding $k\geq3$ logical
qubits, we show that the largest logical group attainable by
automorphisms is generated by all addressable $S$ and $\mathrm{CX}$ gates, and we construct codes attaining it. While this group contains exponentially fewer gates than the full Clifford group, adding one suitable non-Clifford gate yields universality. We further classify the largest logical groups attainable using
qubit permutations, physical single-
qubit Cliffords, or both across general stabilizer and CSS codes, and derive refined bounds for self-dual CSS subclasses. Achieving the maximum-size logical group through
automorphisms requires $n=\Theta(2^k)$ physical
qubits. By contrast, all addressable diagonal Clifford gates, generated by $S$ and $\mathrm{CZ}$, require only $n=\Theta(k^2)$ physical
qubits when implemented using physical single-
qubit Cliffords alone. Both bounds are tight. This
polynomial qubit cost extends beyond Cliffords to all addressable diagonal gates at any fixed level of the Clifford hierarchy, using physical single-
qubit diagonal gates. Thus, for full addressability, the sharpest physical-
qubit cost divide lies between diagonal and $\mathrm{CX}$-type gates, not between Clifford and non-Clifford gates.
Read the full abstract ▾
All authors Jin Ming Koh, Shayan Majidy, Aranya Chakraborty, Anqi Gong, Shi Jie Samuel Tan, Norman Y. Yao
Earth and Planetary Astrophysics
Stellar companions sculpt hot Jupiter formation and spin-orbit evolution
Cheyanne Shariat, Kareem El-Badry, Songhu Wang et al. (+4 more)
Stellar companions can drive hot-Jupiter (HJ) migration and
spin-orbit misalignment, but their role in HJ formation remains uncertain. We construct a homogeneous census of resolved
stellar companions to $147$ northern HJs with measured projected obliquities. We obtain uniform adaptive-optics imaging and combine these observations with
Gaia common proper-motion pairs to identify 8 new companion candidates, bringing the observed companion fraction to $71/147=48\%$. Modeling the full survey selection function yields an intrinsic companion fraction of $62\pm5\%$ for mass ratios $q_\star=0.1$-$1$ and projected separations $s=50$-$50{,}000$~au, roughly 3-4$\times$ enhanced relative to field
stars. Including white-
dwarf companions would increase this fraction further. HJs with resolved companions at $50$-$2{,}000$~au are nearly twice as likely to be misaligned compared to systems without detected companions: $46\%$ compared to $24\%$ ($p=0.009$). The misaligned fraction rises steadily from $5\%$ among the coolest hosts to $80\%$ among the hottest, without a sharp transition at the Kraft Break, while the intrinsic companion fraction remains roughly constant across the temperature range. These trends are consistent with HJs beginning with a broad obliquity distribution, followed by progressively weaker
tidal realignment at higher
stellar temperatures. Contrary to previous work, we find that most
stellar companions in our sample are capable of driving eccentric Kozai-Lidov (EKL) oscillations to the
tidal limit under suitable orbital configurations, making high-eccentricity migration dynamically promising. Taken together, these results indicate that
stellar companions sculpt HJ formation and
spin–orbit architectures.
Read the full abstract ▾
All authors Cheyanne Shariat, Kareem El-Badry, Songhu Wang, Xian-Yu Wang, Malena Rice, Jerry W. Xuan, David R. Ciardi
Solar and Stellar Astrophysics
Precise and Accurate Mass and Radius Measurements of Thirty Galactic Metal-Poor Stars in Detached Eclipsing Binaries
D. M. Rowan, K. Z. Stanek, C. S. Kochanek et al. (+4 more)
Detached eclipsing binaries have long served as fundamental
benchmarks for making direct comparisons with
stellar models. By combining the light curve and radial velocity orbit of a detached eclipsing binary, precise masses and radii can be measured for both components. However, catalogs of eclipsing binaries are dominated by
solar-
metallicity stars. Here we characterize fifteen eclipsing binaries selected to have $[\rm{M/H}]\lesssim -0.5$ using TESS photometry,
Gaia XP spectra, large spectroscopic surveys, and
Galactic kinematics. We use three spectrographs, PEPSI on the Large Binocular
Telescope, the Levy spectrograph on the Automated
Planet Finder
telescope, and CHIRON on the SMARTS 1.5m
telescope. We jointly model the light curves and radial velocities and measure the
stellar masses and radii with median fractional uncertainties of < 1%. We use spectral disentangling to extract the spectra of both
stars and model them using the spectral synthesis code Korg. We find three systems with $[\rm{M/H}] < -2.0$, two of which are kinematically consistent with the Milky Way's
halo. J0001+2156 has $[\rm{M/H}]=-2.51$, making it the most metal-poor detached eclipsing binary characterized to date. Eleven of the twelve other systems have $[\rm{M/H}] \lesssim -0.5$ and are split between being highly ($[\alpha/\rm{M}] > 0.25$) and moderately ($[\alpha/\rm{M}] < 0.25$) $\alpha$-enhanced. We compare our measured masses and radii to predictions from theoretical isochrones and estimate the ages of the binaries. We estimate ages with typical uncertainties of $\lesssim 0.5$ Gyr, and three of the binaries have estimated ages greater than 10 Gyr, up to $13.8^{+0.8}_{-0.7}$ Gyr. Finally, we update the empirical relations for
stellar masses and radii on the main sequence using a sample of 621
stars.
Read the full abstract ▾
All authors D. M. Rowan, K. Z. Stanek, C. S. Kochanek, Benjamin J. Fulton, Ilya Ilyin, Howard Isaacson, David V. Martin
High Energy Physics (Theory)
Vijay Balasubramanian and Charlie Cummings
We derive an
entropy formula for a recently proposed tensor network model of states in 3D
gravity which satisfy the constraint equations of diffeomorphism invariance, and including the effects of matter. After accounting for the invertibility of the spacetime metric, we argue that this formula matches the
quantum extremal surface formula (for small but finite $G_N$), and is not restricted to time symmetric spacetimes. We derive the result via canonical quantization without relying on path integrals, and discuss an extension to compact bulk subregions.
Read the full abstract ▾
Astrophysics of Galaxies
Rubin LSST DP2 unveils almost-dark galaxies in the Virgo Cluster
Minh Ngoc Le, Johan H. Knapen, Željko Ivezić et al. (+26 more)
Galaxies with the faintest surface brightness are currently known only in the Local Group. Similar objects should exist beyond our vicinity and are crucial for understanding
galaxy evolution, structure, and dark matter content, yet surveys have not reached the depth required to detect them systematically. We present a population of seven almost-dark
galaxies identified in Data Preview 2 of the Rubin Legacy Survey of Space and Time. They surround M49 in the vicinity of the Virgo Cluster, and exhibit central surface brightnesses in the range of $26.9 - 28.5 \, \mathrm{mag \, arcsec^{-2}}$ in the $g$-band, with half-light radii of $0.6 - 4.6 \, \mathrm{kpc}$ at the distance of Virgo, and
stellar masses of $10^6 - 10^7 \, \mathrm{M_\odot}$. Their characteristics are analogous to those of the faintest and low-mass
galaxies identified among satellite
galaxies And XXI, And XXIII, and And XXV in the Local Group. This discovery demonstrates the power of the forthcoming Rubin LSST 10-year survey to uncover extremely faint
galaxies at scale, promising the large statistical samples needed to constrain the faint-end luminosity function and the nature of dark matter.
Read the full abstract ▾
All authors Minh Ngoc Le, Johan H. Knapen, Željko Ivezić, Junais, Aaron Watkins, Dan S. Taranu, Pierre-Alain Duc, Anthony Englert, Erfan Nourbakhsh, J. Anthony Tyson, Jiaxuan Li, Roberto J. Assef, Uzay Aydin, Jose Benavides, Natanael M. Cardoso, Priyanka Chakraborty, Ricardo Demarco, Bililign Dullo, Alister W. Graham, Marc Huertas-Company, Kian-Tat Lim, Jakub Nadolny, Dieu Nguyen, Reynier Peletier, Rossella Ragusa, Kevin Reil, R. Michael Rich, Hector Hernandez-Toledo, Michael H. F. Wilkinson
Combinatorics
A Complete Proof of the Strong Conjecture about $F$-Irregular Graphs
Tatiana Dovzhenok and Artem Filuta
A graph $G$ is called $F$-irregular if all its vertices have distinct $F$-degrees, defined as the number of subgraphs of $G$ isomorphic to a given graph $F$ and containing the respective vertex. We prove the Strong
Conjecture about $F$-irregular graphs (Dovzhenok, Filuta, and Chuhai, 2024), which states that for every connected graph $F$ of order at least three, there exist infinitely many $F$-irregular graphs. Fundamentally generalizing the classical existence
conjecture by Chartrand et al. (1987), this work presents an authorized English translation of our original February 2024 manuscript, which was publicly presented in full at two scientific conferences the same year.
Read the full abstract ▾
Artificial Intelligence
Characterizing Web Search by Conversational LLM Agents: From Search Decisions and Strategies to Results and Responses
Mahsa Amani, Seungeon Lee, Abhisek Dash et al. (+9 more)
Conversational LLM
agents increasingly rely on Web search, yet the end-to-end lifecycle of agentic search remains poorly understood. We present the first study of Web search across four major conversational platforms (ChatGPT, Claude, Grok, and DeepSeek), combining real-world user interactions (invivo) with controlled experiments using the same platform's models by their APIs (invitro). We investigate the quality of agentic decisions to invoke Web search, their strategies to formulate queries, the potential domain preferences in the search results they receive, and the choices they make when transforming search results into grounded responses. We find that Web-search decisions vary substantially across platforms and models, while more frequent Web-search invocation does not necessarily yield better response quality. We further show that conversational
agents employ different complex querying strategies and that platform specific search engines return search results from their preferred domains. Finally, although responses are largely grounded in search results, some claims rely on uncited search results, raising concerns about attribution and reliability. Our findings have important implications for the design of future AI
agents and Web search tools optimized for conversational
retrieval.
Read the full abstract ▾
All authors Mahsa Amani, Seungeon Lee, Abhisek Dash, Asmaa El Fraihi, Yunah Jang, Elisabeth Kirsten, Qinyuan Wu, Krishna P. Gummadi, Manish Gupta, Abhilasha Ravichander, Muhammad Bilal Zafar, Soumi Das
Machine Learning
Randomized SVD Approximations for Spectral Co-Clustering of Word-Document Matrices
Fateme Mazdarani and Carlos Toxtli
Spectral co-clustering is a useful tool for discovering latent structure in word-document
matrices, but its reliance on singular value decomposition (SVD) can make standard formulations expensive on high-dimensional data. This paper presents two randomized approximations for normalized spectral co-clustering of bipartite text data when the numbers of document and word clusters may differ. The first method uses randomized SVD through random projection, while the second combines partial SVD with element-wise random sampling. Across real-world and synthetic
datasets, both methods reduce runtime relative to the full-SVD baseline, but their behavior depends on
matrix sparsity. The random projection method is the more reliable approximation across the tested settings, whereas the sampling-based method is most useful on denser
matrices and provides limited benefit on already sparse text data. These results show that randomized approximations for spectral co-clustering should be selected according to the underlying structure of the data.
Read the full abstract ▾
Machine Learning
Block Parallelism For Efficient Distributed Long-Context Diffusion Language Model Training
Tarun Suresh, Pranshu Chaturvedi, Hangoo Kang et al. (+4 more)
Block
diffusion language models (BDLMs) combine autoregressive dependencies across blocks with parallel denoising within blocks, but long-context training is constrained by distributed
attention communication and activation
memory. Conventional context parallelism (CP) shards the combined clean-plus-corrupted sequence by position, communicating shared clean K/V together with block-specific corrupted K/V and their gradients. We observe that the BDLM objective separates over target blocks. We introduce block parallelism (BP), a new distributed parallelism dimension that assigns each corrupted-block computation to one rank. To scale BP to long contexts, we introduce context-sharded block parallelism (CSBP), which also shards the shared clean sequence across those ranks. CSBP keeps corrupted K/V and gradients local, avoids replicated clean prefixes, and preserves BDLM training semantics. On 16 H200 GPUs at 256K context, CSBP improves throughput over the best baseline by 1.18-1.45x for supervised fine-tuning and 1.27-1.33x for conversion of autoregressive models to BDLMs, while matching or reducing peak HBM. Full-model speedup reaches 1.61x at 512K. On eight H100 GPUs, CSBP accelerates DFlash2 speculative-decoder training by 2.48x at 512K and 7.59x at 1M. In matched 12-hour DiffusionGemma 26B-A4B SFT runs, CSBP achieves higher pass rates at every trained checkpoint on SWE-bench Verified and Terminal-Bench Lite. Code: https://github.com/ScalingIntelligence/Turbo-dLLM
Read the full abstract ▾
All authors Tarun Suresh, Pranshu Chaturvedi, Hangoo Kang, Parth Shroff, Ishan S. Khare, Hermann Kumbong, Azalia Mirhoseini
Quantum Physics
Lazy training of quantum physics informed neural networks
Anderson Melchor Hernandez and Giacomo De Palma
We study the gradient-flow training dynamics of
quantum physics-informed neural networks (QPINNs) for the solution of second-order
elliptic partial differential equations with Dirichlet boundary conditions. We consider parameterized
quantum circuits as function approximators and analyze their overparameterized regime through the lens of the neural tangent kernel (NTK). Our contribution is a nonasymptotic lazy-training theory for QPINNs and their variational formulation: we prove that, for sufficiently large
circuit width, the nonlinear gradient flow is quantitatively approximated by a linearized NTK model, with explicit bounds depending on the number of
qubits,
circuit depth,
circuit light-cone geometry, and the dimension of the domain of the solution to the PDE.
Read the full abstract ▾
Computer Vision
RAUL: Reference-Assisted Ureteroscopy Localization for Skill Assessment
Fangjie Li, Mai Bui, Charan Mohan et al. (+4 more)
Objective: Incomplete
navigation of anatomy during ureteroscopic kidney stone surgeries can contribute to repeat interventions. While skilled surgeons have lower reintervention rates, there are no objective metrics to quantify scope-
navigation performance to evaluate when a trainee becomes skilled. This work aims to recover ureteroscope trajectories from endoscopic
video and derive
navigation metrics to quantify differences in skill. Methods: We propose RAUL, a reference-assisted reconstruction framework for recovering ureteroscope trajectories from ureteroscope
videos only in phantoms. For each phantom, we use a slow, high-quality reference exploration
video to generate a reference reconstruction. We localize subsequent exploration
videos against this reference. We evaluate localization accuracy against electromagnetically tracked scope pose. We compute
navigation metrics from phantom exploration trajectories to compare
surgical residents across experience levels. Results: The proposed reference-assisted framework achieves a mean translation root mean square error of $0.5 \pm 0.1$ mm across 9 phantoms. Compared to standard Structure-from-Motion (SfM), the proposed pipeline increases frame-wise localization coverage from $50.5 \pm 14.9\%$ to $86.1 \pm 7.2\%$ of all
video frames. The reconstructed trajectories revealed significant differences between high- and low-experience trainees in established
navigation metrics. Conclusion: RAUL enables substantially more complete recovery of ureteroscope trajectories from
videos compared to standard SfM pipelines, enabling trajectory-based skill assessment without additional tracking equipment. Significance: To the best of our knowledge, this is the first use of
video-only recovery of ureteroscope trajectories without external tracking sensors for skill assessment, supporting scalable automated assessment of ureteroscopy
navigation skill.
Read the full abstract ▾
All authors Fangjie Li, Mai Bui, Charan Mohan, Michael Miga, Matthieu Chabanas, Nicholas Kavoussi, Jie Ying Wu
Cosmology and Nongalactic Astrophysics
Where the Forest Goes Dark: Halo-centered Characterization of High Column Density Systems with IllustrisTNG
Victor R. Soares da Silva, Daniel López-Cano, L. Raul Abramo et al. (+2 more)
High column density systems of neutral hydrogen (HCDs) — Lyman limit systems (LLSs), sub-damped and damped Ly$\alpha$ absorbers (DLAs) — contaminate both the 3D and the 1D statistics of the Ly$\alpha$ forest. Most DLAs are masked out from the analyses, but cleaning
algorithms are not perfect, and weaker LLSs are individually undetectable, so the residual contamination is absorbed into nuisance parameters that dilute the cosmological constraining power. Simulations provide the ideal setting to study this effect, allowing absorption features to be directly traced back to the gas producing them. We identify high-column-density structures on the gas
cells of the TNG50 hydrodynamical simulation, deblend them in velocity space, and split every sightline into HCD-only and forest-only spectra. Applied to
halos at $z \simeq 3$ out to 50 virial radii, it yields each class's covering fraction against impact parameter $b$ and
halo mass. The impact parameter defines the type of absorption appearing in the spectrum: DLAs give way to sub-DLAs at $b \approx 0.11\,R_{200c}$, sub-DLAs to LLSs at $0.3$, and LLSs to the Ly$\alpha$-forest at $0.5$, which covers $\simeq 78\%$ of sightlines at $R_{200c}$. In units of $R_{200c}$ the sequence is nearly mass-independent over three decades in mass, so the virial radius sets the scale of the neutral gas distribution. A single Voigt component recovers the column density of essentially every damped system, but not of LLSs, whose absorption features often arise from several separate contributions from gas structures along the sightline. Our detailed characterization of HCDs gives the first step to the construction of advanced techniques to directly forward-model their contribution to Ly$\alpha$ spectra.
Read the full abstract ▾
All authors Victor R. Soares da Silva, Daniel López-Cano, L. Raul Abramo, J. Chaves-Montero, Francisco Maion
Computers and Society
The AR Fairness Metamodel: A Structured Framework for Fairness Measures
Julian Alfredo Mendez and Timotheus Kampik
This paper presents the AR
fairness metamodel, a framework designed to represent, analyze, and compare different
fairness scenarios. The metamodel considers key elements, such as
agents, resources, and their attributes, and enables the systematic definition and comparison of various
fairness measures. We provide examples involving both discrete and continuous measures, including equality, equity, group
fairness, individual
fairness, the Gini index, the Theil index, Jain's
fairness index, and a detailed
fairness measure for Australia's Child Care Subsidy. We also explore relationships among group
fairness, individual
fairness, and envy-freeness, supported by formal proofs. At the conceptual modeling level, our approach builds on the Tiles framework, which offers modular components that can be connected to capture diverse
fairness definitions. The goal is to make AR-based
fairness definitions practical and adaptable across contexts, providing a clear way to define, compare, and evaluate them. An implementation of the Tiles framework is available as an open-source tool, and can support
fairness modeling and evaluation across a wide range of applications.
Read the full abstract ▾
Human-Computer Interaction
”Bless his heart... he thought all we did was push a button": Understanding Worker Challenges with U.S. Election Technology
Delaney Gomen, Josiah Hester, Naveena Karusala et al. (+1 more)
This paper explores how the technologies U.S. election officials depend on also challenge their work. Every election, misleading narratives stem from errors that occur while using election technology, threatening election official safety and eroding faith in elections. We seek to understand the impact of these challenges directly from worker perspectives. This paper presents a qualitative analysis using combined interview and survey data from a total of 50 election officials representing 20 states. Our findings present design as a substantial factor in mistakes when using election technology, but other obstacles, like lack of funding and technical support, also strain operations. We show how election technology design can lead to harms for election officials and connect those harms to their negative effects on democracy. Our work emphasizes the potential for researchers in human-centered computing to support U.S. election officials-and democracy-by working towards better usability across election technologies.
Read the full abstract ▾
All authors Delaney Gomen, Josiah Hester, Naveena Karusala, Michael Specter
Functional Analysis
Gabor Frame Regions and Non-Frame Obstructions for an even Rational Window
Riya Ghosh
We study the Gabor frame properties of the rational window $$ g(x)=\frac{x^2-1}{(x^2+1)(x^2+4)(x^2+9)}, $$ whose poles occur in symmetric pairs. We prove that every
lattice satisfying $0<\alpha\beta<1/3$ generates a frame, and we establish an additional frame region for $\beta\ge1$ and $1/3\le \alpha\beta<0.47373$. Furthermore, we show that the rational hyperbolas $\alpha\beta=\frac{p}{3p-1}$, for $p\ge2,$ are entirely contained within the frame set. In contrast, we construct explicit non-frame
lattice points on the hyperbolas $$\alpha\beta\in\left\{\frac{1}{3},\frac{1}{2},\frac{2}{3},\frac{3}{4}\right\}.$$ Finally, for the density family $\alpha\beta=p/(p+1)$, we derive a
symmetry reduction of the associated Zibulski-Zeevi
matrix, providing numerical evidence for a richer structure of non-frame obstructions.
Read the full abstract ▾
Combinatorics
Intractable enumeration problems are like Russian nesting dolls: structural properties of monomer-dimer coverings on two-dimensional quadratic lattices
Yong Kong
Counting the number of coverings of $s$ dimers on two-dimensional quadratic
lattices is considered as intractable and belongs to \#P-complete class. We reveal the structure of the exact solution to the problem and provide an explicit formula for it, which includes $s-1$ nesting sums. This results in an exponential time complexity of $O(2^s)$. The solution is explicitly determined by a sequence that exhibits double-exponential growth.
Read the full abstract ▾
Computer Vision
Open ultrasound foundation model for robust segmentation and clinical measurement across heterogeneous settings
Chao Qin, Fahad Shahbaz Khan, Salman Khan et al. (+4 more)
Ultrasound is the most widely deployed imaging modality worldwide, yet clinical AI remains fragmented into narrow single-task models that fail when device, operator, or anatomy changes. Here we present SonoCorpus, an open resource unifying 456,963 images and 1,626,085 expert masks from 53 public
datasets spanning 24 clinical applications and 17 countries, and SonoBase, an interactive
segmentation foundation model pretrained on it. Across fifteen evaluation
datasets introducing new organs, devices, operators, and geographies, SonoBase outperforms SAM2, MedSAM2, and the concept-promptable MedSAM3 on every
dataset and matches per-
dataset specialist models trained on the same data; on fully external data it exceeds the accuracy these baselines achieve on their own in-distribution
benchmarks. Ejection fraction derived from its
segmentations falls within inter-observer variability (6.63% error), with fewer misclassifications at the defibrillator-candidacy threshold than either promptable baseline (13% versus 18–42%); fetal head-circumference (1.81~mm) and gestational-age (1.2 days) errors fall below inter-observer variability. Where a baseline fails outright, one in four test cases, SonoBase recovers a usable
segmentation in 81% of them, including on handheld probes operated by minimally trained users in two low- and middle-income countries (Sierra Leone and Tanzania). Five labeled examples can help the model adapt to a new setting, and the identical training protocol transfers well to newer models such as SAM3, locating the advantage in ultrasound-specific pretraining rather than any single architecture. To ensure reproducibility and enable the community to build on SonoBase as a platform, we release all checkpoints, optimizer states, data-split indices, deduplication hashes, and starter code at https://github.com/AlfredQin/sonobase.
Read the full abstract ▾
All authors Chao Qin, Fahad Shahbaz Khan, Salman Khan, Sarim Ather, Siddiq Anwar, Rao Muhammad Anwer, Shadab Khan
Materials Science
Melting of corundum ({\alpha}-Al2O3) under compression to 8 GPa: A decreasing melting slope
Vladimir L. Solozhenko
Melting of corundum ({\alpha}-Al2O3) has been studied under compression up to 8 GPa using in situ electrical resistivity measurements. The melting temperature increases monotonically from T0 = 2317 K at ambient pressure to 3090 K at 7.7 GPa. Fitting the experimental data to the Simon-Glatzel equation yields the empirical parameters a = 5.60 GPa and c = 0.334, providing a closed-form analytical description of the entire melting boundary. Most significantly, the melting slope, dTm/dp, decreases continuously from 138 K/GPa at ambient pressure down to 78 K/GPa at 7.7 GPa, confirming unequivocal negative
curvature of the melting curve (d2Tm/dp2 < 0). This phenomenon is consistent with the preferential densification of the
liquid phase and the associated reduction in the volume change upon melting as pressure rises. These findings provide a robust empirical standard for the high-pressure melting of corundum, establishing essential constraints for future ab initio simulations, thermodynamic databases, and geophysical models of alumina-rich deep-Earth systems.
Read the full abstract ▾
Robotics
Grasping by interconnection: robust closing motions from coarse object templates
Julien Vanderheyden, Guillaume Drion, Fulvio Forni et al. (+1 more)
Dexterous
robot hands must often grasp objects whose shape, size, and pose are known only approximately. Grasp planners typically require accurate object models or correct errors with feedback, but how much inaccuracy a closing motion can tolerate on its own remains unclear. To address this question, we designed a motion planner based on four principles: a coarse template of the object, human grasp types, an object-centric interaction, and compliant, sliding contacts instead of prescribed contact points. This paper presents the planner, implemented through virtual model control, and its evaluation on a Shadow Dexterous Hand. Without feedback, the planned closing motions tolerated size errors of about 1cm and pose errors of several centimeters and tens of degrees, a wider range than a state-of-the-art data-driven planner in 25 of 27 tested conditions. They also grasped 82.5% of 80 everyday objects and succeeded within an autonomous pipeline. Robustness can thus be designed into the closing motion itself, rather than left only to feedback. This planner opens a path toward reliable manipulation in uncertain settings, which we will pursue by combining it with adaptive feedback control on the physical hand.
Read the full abstract ▾
All authors Julien Vanderheyden, Guillaume Drion, Fulvio Forni, Pierre Sacré
Statistical Methodology
Binning-Independent Bayesian Analysis of Time-Dependent Perturbed Angular Distribution data
Franziskus von Spee
We propose a new approach for analyzing Time-Dependent Perturbed Angular Distribution data. In this, a likelihood is constructed with the help of conditional probabilities of single events and binning of time data is avoided. This likelihood is used in a
Bayesian framework to calculate the posterior probability density function for the $g$ factor of interest. This approach is compared to more traditional approaches that use binned data by analyzing simulated
datasets. In many cases the resulting posterior probability densities are observed to be
multimodal and the results can thus often not be summarized with a single Gaussian approximation. We find that results from the new approach are more reliable for low-statistics
datasets.
Read the full abstract ▾
Cryptography and Security
PAPC: Platform Mediation for Privacy-Propagation Externalities in AI-Mediated Workflows
Tao Huang, Guosen Wu, Chen Hou et al. (+1 more)
AI-mediated platforms coordinate work through LLM
agents acting for different principals. In these workflows,
privacy loss can be created before a final answer appears: a
memory write, shared-workspace update, inter-
agent message, or tool event may impose downstream exposure cost on another principal. We model this failure mode as a
privacy-propagation externality, where the cost of a raw disclosure depends on
topology and fanout as well as content. We present PAPC, a platform-mediated mechanism that intercepts information-moving events before they update shared state or external channels. PAPC combines policy, provenance,
topology/fanout, privilege, and content signals to allow an event, release a policy-safe abstraction, quarantine raw content, block a transition, or narrow onward rights. The model explains why final-output control misses intermediate exposure costs and why high-fanout objects amplify propagation. Across
retrieval-
memory and multi-
agent workflow
benchmarks, PAPC preserves deterministic task completion and eliminates measured exact raw-value and external raw-value exposure. The results position event-level mediation as a platform-governance primitive for
agent-mediated online work.
Read the full abstract ▾
All authors Tao Huang, Guosen Wu, Chen Hou, Guolong Zheng
Computers and Society
Making Local Government Contracts Legible: A Computational Pipeline for Classifying and Mapping Intergovernmental Service Agreements
Mohsen Ghasemizade, Raúl Gutiérrez-Meave, Cailin Gramling et al. (+4 more)
Interlocal agreements are one of the primary instruments through which local governments formalize collaboration for public service delivery, yet the institutional and financial content encoded in these contracts has remained inaccessible to systematic analysis at scale. This paper introduces an end-to-end computational pipeline for classifying intergovernmental agreements by institutional form and extracting financial relationships between principals and
agents in service contracts. Applied to Iowa's 28E archive (N = 21,629), the largest
dataset of interlocal agreements in the United States, the pipeline combines LLM-based summarization and classification across LLaMA 3.1, GPT 5.2 Pro, and Gemini 3 Pro on a four-class classification task that distinguishes agreements as either service contracts, resource sharing agreements, joint operations agreements, or new joint entity agreements. We also identify the financial principal and
agent in these agreements and contracts, as well as the resulting dollar amounts and represent them on a directed network. The resulting financial network is organized around a small number of dominant service providers, with counties serving as the most structurally versatile actors, and cities as predominantly principals. By rendering the content of Iowa interlocal agreements analyzable at scale for the first time, this pipeline establishes a reusable methodology that researchers and state agencies can apply to track how public dollars move across local governments and to identify entities that depend heavily on a small number of providers.
Read the full abstract ▾
All authors Mohsen Ghasemizade, Raúl Gutiérrez-Meave, Cailin Gramling, Aviral Chawla, Michael Robinette, Kate Albrecht, Juniper Lovato
Analysis of PDEs
Long-time dynamics for the magnetic Schrödinger equation on the torus
Julien Lechaux
We investigate time-dependent semiclassical measures associated with the
magnetic Schrödinger flow on the flat two-dimensional
torus in long-time regimes. Under a
geometric nonvanishing assumption on an effective
magnetic force, we identify two thresholds and prove a dynamical form of
quantum unique ergodicity for such systems. If the time scale $\tau_h \gg h^{-1/2}$, then the configuration marginal is the normalized Lebesgue measure on $\mathbb T^2$. In addition, if $\tau_h \gg h^{-1}$, then the momentum marginal is, conditionally on each regular level set, the normalized invariant measure on that curve.
Read the full abstract ▾
Statistical Methodology
When a High Score Is an Illusion: Certifying Genuine versus Repackaged Forecasting Skill
Pin Ni, Francesca Medda and Ramin Okhrati
Ranks depend on the observations used for comparison. Reusing those observations can add association between forecast and outcome rank contrasts even when the evaluated forecast and outcome stay fixed. We characterize assignments that preserve association between specified population-rank contrasts, including forecast rank minus baseline rank compared with outcome rank minus baseline rank. Conditional on independent training, whole trajectories are sampled independently from a common law, with unrestricted dependence within each trajectory. The expected score separates into its target and an explicit interaction between map pairs. When reassigning references, we keep the learned maps, reference law and coefficient row sums fixed. Zero weighted reference overlap for every map pair is necessary and sufficient for preservation uniformly over permitted maps and laws. An unbiased three-trajectory kernel estimates the interaction; independent evaluation and validation provide finite-sample lower bounds. Complete U-statistics estimate the same target directly when all draws can be recombined. Sharp shared-baseline ranges, including ties, tighten both constructions. In a Beijing air-quality archive, interaction accounts for 91.4% to 94.5% of seven learned forecasts' expected shared scores under the empirical archive law. Separate results address category-fitting error and temporal feedback. In a matched category-control null experiment, distinct references reduce rejections from 402 to 41 out of 1,000 panels.
Read the full abstract ▾
Probability
TV between Bernoulli products, up to constants
Gleb Smirnov
We obtain upper and lower bounds on the total variation distance between two product Bernoulli measures. The bounds are efficiently computable and match up to an absolute constant factor.