Differential Geometry
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Thursday 10 September
We introduce and study a scalar $q$-deformation of the cross-ratio on $\mathbb P^1(\mathbb Q)$. Our construction is based on the notion of $q$-deformed rational numbers due to Morier-Genoud and the author. The $q$-cross-ratio is invariant under $\mathrm{PSL}(2,\mathbb{Z})$, while elements of determinant $-1$ of $\mathrm{PGL}(2,\mathbb{Z})$ act by $q\mapsto q^{-1}$. A principal result is its relation to $q$-deformed Coxeter friezes associated with rational polygons. The expansion at $q=e^h$ yields an algebraically independent sequence of modular invariants and relative invariants, although this sequence does not separate modular orbits. We compute the first two nonconstant coefficients of this expansion explicitly.
Systems and Control
Structured Stochastic Representations of Integrated Dynamic Strategies
Dynamic allocation decisions couple present resource use to evolving internal conditions, delayed returns, and future costs. We represent this interaction by four probability localizations linked through regime-indexed, graph-constrained column-stochastic operators. Pre-action state or context selects a locally affine model, while action-dependent changes update subsequent regimes, yielding a causal switched representation of nonlinear evolution. We characterize operator identifiability relative to the graph, the stochastic constraints, and the sampled embedding, separating coefficient recovery from predictive equivalence on the decision domain. Decision making is then formulated through implementable return–cost acceptability regions. Finite-horizon error propagation supplies conservative classification margins, and simultaneous intervals distinguish model-relative near-optimality from certified $\epsilon$-optimality over a declared finite policy class. Regime-indexed stochastic feedback is admitted when it satisfies the same certification test. Reproducible synthetic laboratories for personal preparation, supplier participation, and customer retention illustrate exact, operator-supplied, and noisy feedback cases. Multinomial experiments show improving recovery of the feedback function and fewer unresolved decisions with increasing sample size, while unrestricted off-policy recovery remains limited. The contribution is a structure-preserving representation–identification–decision workflow, not a domain-specific physiological or commercial calibration.
Read the full abstract
Solar and Stellar Astrophysics
Science operation, data handling, and ground support system of the SOLAR-C mission
SOLAR-C is an international solar-observing satellite mission led by Japan Aerospace Exploration Agency (JAXA). It aims to elucidate mass and energy transport in solar atmospheres through extreme ultraviolet (EUV) spectroscopy. The mission carries the EUV High-throughput Spectroscopic Telescope (EUVST) and the Solar Spectral Irradiance Monitor (SoSpIM), enabling comprehensive observations across a wide temperature range (10^4 K to 10^7 K) with minimum temperature gaps and high spatial and temporal resolution. To achieve its science objectives regarding atmospheric heating and solar flare eruptions, SOLAR-C implements a flexible and responsive operational procedure and a data processing system, while building on the heritage of the Hinode and IRIS satellites. The Chief Observer creates observation timelines that include core observation plans and approved proposed plans while taking into account the solar activity levels. A SpaceWire communication architecture is employed onboard the spacecraft, and a method is implemented in which the mission instrument temporarily acts as the network master during data transfer to support the high data rate requirements. Telemetry is downlinked at ground stations worldwide and gathered at the Institute of Space and Astronautical Science, JAXA. The EUVST data are calibrated at the SOLAR-C Science Center at Nagoya University, while the SoSpIM data are calibrated at the Processing and Archiving Facility before being integrated into the science data products. The data will be made publicly available immediately. The integrated operational scheme for this mission is expected to advance our understanding of solar atmospheric heating and flare processes.
Read the full abstract
Computation and Language
Rethinking Verbalized Confidence for LLM-as-a-Judge: A Compatibility Shift on Post-2025 Proprietary Models
Verbalized confidence, long dismissed as overconfident, coarse, and prone to round-number clustering, is now the more robust soft-scoring mechanism for LLM-as-a-Judge on top-tier proprietary models. Across SummEval, AggreFact, and HelpSteer2, spanning up to 18 LLMs, we show that the standard advice to prefer log-probabilities no longer holds on post-2025 models, where verbalized confidence is the better signal. We call this a compatibility shift. On top of a standard verbalized-confidence baseline, we introduce two new ingredients: an overconfidence advisory and self-debate. Together they improve calibration, score-distribution spread, and robustness to task subjectivity. We further observe a generation effect: post-2025 models accommodate these two additions with little balanced-accuracy cost, whereas pre-2025 models pay a measurable penalty. Compared with logprob-based G-Eval, verbalized confidence is the more subjectivity-robust soft signal on GPT-family top-tier releases. The shift is invisible under accuracy-only reporting. Rather than defaulting to hard predictions, we recommend broader use of soft scoring in LLM-as-a-Judge. More broadly, verbalized confidence has moved from a weaker substitute for logprobs to a practical soft-scoring mechanism for contemporary LLM judges.
Read the full abstract
Materials Science
U-Shaped Fermi-Level Dependence of Point-Defect Aggregation Enthalpy in Silicon
Point defects govern the electronic and structural behavior of silicon, yet their configuration and energy depend on both the charge state and the Fermi level (EF). Using an unbiased global structure search with first-principles calculations, we screened more than 1,500 candidate structures, covering intrinsic defects and C, H, O, N, P, and B impurities in charge states from -2 to +2. The reaction enthalpy of defect aggregation depends on EF in a U shape, being strongest where the net charge transferred during the reaction vanishes and weaker toward both band edges. This U arises because complexes are more charge-neutral than their isolated constituents, so the net charge transfer changes sign across the gap. The point where this sign change occurs is set by the transition levels of the specific defects. For typical reactions, the driving force is tunable by 0.3-0.5 eV, which shifts the equilibrium complex concentration by five to eight orders of magnitude at room temperature. Our results make the Fermi level a practical lever for defect engineering in as-grown and irradiated silicon.
Read the full abstract
Machine Learning
Importance Weighting for Unlabeled-unlabeled Learning under Distribution Shift
Unlabeled-unlabeled (UU) learning allows us to learn a binary classifier from two sets of unlabeled data with different class-priors. It is a general framework because it includes a wide variety of supervised learning such as positive-unlabeled (PU) learning, noisy label learning, and similarity-based learning. Existing UU learning assumes that the test and training distributions have the same class-conditional densities. However, this assumption rarely holds in practice due to distribution shifts. This paper proposes a distribution shift adaptation method for UU learning that uses UU data in the training distribution and a few UU data in the test distribution. The proposed method is based on the importance weighting, which minimizes the test risk by using training data with estimated importance weights. Although existing importance weighting methods cannot handle UU data, we show that it can be done in a principled manner. Thanks to the generality of UU learning, our method can handle various learning problems such as PU and noisy label learning under distribution shift within a single framework while existing methods are usually tailored to a specific problem. Moreover, it does not require any assumption of the shift types such as covariate shift. We experimentally demonstrate the effectiveness of the proposed method with real-world datasets.
Read the full abstract
Computation and Language
Distribution-aware Language Neuron Identification in Multilingual Large Language Models
Multilingual large language models (mLLMs) contain a small fraction of feed-forward neurons that are sensitive to particular languages, commonly termed language-specific neurons. Existing work measures language specificity using the entropy of each neuron's language-wise probabilities of being active, where a neuron is considered active when its activation value is positive. However, this approach may not fully capture the multilingual nature of mLLMs, where language representations are distributional and mutually related. We propose Distribution-aware Language Neuron selection, which leverages pairwise relationships between per-language activation distributions over the full activation range, including negative values. Specifically, we quantify each neuron's language specificity by clustering languages using pairwise overlap coefficients between their activation distributions. Across two mLLMs and two held-out corpora, our identifier more effectively isolates language-specific causal effects, yielding up to 4.9$\times$ higher on-target language damage per neuron while preserving off-target language performance.
Read the full abstract
Artificial Intelligence
Demystifying the Privacy-Utility Trade-off in LLM Interactions
The integration of Large Language Models into daily tasks relies on context-rich instructions, inevitably exposing sensitive user information. Current privacy-preserving methods typically employ context-agnostic static rules, causing severe utility degradation. However, the specific mechanisms governing how sanitization impacts downstream performance remain largely underexplored. To address this, we conduct a systematic analysis to deconstruct the privacy-utility trade-off, uncovering three underlying mechanisms: (1) Context-Dependent Utility, which first establishes when to sanitize by revealing that data value shifts from critical constraints to dispensable noise based on user intent; (2) Strategic Adaptation, which subsequently determines how to sanitize by dictating that the choice between removal and replacement depends on the task's reliance on factual integrity versus structural coherence; and (3) Combinatorial Interplay, which finally extends the protection scope by demonstrating that attributes form a semantic web of synergistic dependencies or antagonistic redundancies. Guided by these insights, we introduce an intent-driven local protection framework. By distilling a lightweight model Veilmind-4B to drive a dynamic extraction-sanitization-restoration pipeline, our approach reaches a low-leakage privacy point while preserving substantially higher response utility than existing privacy-oriented baselines, advancing the privacy-utility trade-off toward the Pareto frontier.
Read the full abstract
Cosmology and Nongalactic Astrophysics
Two-parameter continuous deformation of Starobinsky inflation as a bridge between Planck and ACT DESI data with $N_\star\in(50,60)$
We present a family of plateau-type inflationary potentials, eq.~\eqref{Vgeneral}, and analyze a two-parameter $\alpha\beta$-Starobinsky specialization that interpolates continuously between a maximal plateau ($V\!\to\!V_0$) and a submaximal plateau ($V\!\to\!V_\infty<V_0$), where $V_0$ is the overall scale of the potential. In the $\beta\!\to\!0$ limit our framework reproduces the $\alpha$-Starobinsky predictions $n_s\simeq 1-2/N_\star$ and $r\simeq 12\alpha/N_\star^2$. For $\beta>0$ with $x_\star\gg 1/\beta$ the slow-roll scaling laws change to $n_s\simeq 1-\frac{4}{3N_\star},\, r\simeq\mathcal{C}(\alpha,\beta)\,N_\star^{-4/3},$ with an explicit coefficient $\mathcal{C}(\alpha,\beta)$ set by the plateau truncation. This deformation lifts $n_s$ at fixed $N_\star$ while further suppressing $r$, reconciling the Planck~2018 constraint $n_s=0.9649\pm0.0042$ (68% CL) and BICEP/Keck18 data $r_{0.05}<0.036$ (95% CL), with the higher central values $n_s\sim0.97$–$0.98$ preferred by ACT+DESI~DR2 (BAO), within the theoretically motivated interval $N_\star\in(50,60)$ and without exotic reheating. We provide an exact identity for $V/V'$ enabling analytic control of $N_\star$, a practical crossover criterion $\beta\,x_\star\ll1$ vs.\ $\gg1$, and a transparent mapping between $(\alpha,\beta)$ and the observables $(n_s,r,N_\star)$. These yield sharp, testable signatures, particularly the softened $N_\star$-scaling of $r$, that distinguish a maximal from a submaximal plateau with upcoming CMB and LSS data.
Read the full abstract
Solar and Stellar Astrophysics
Classical Nova V1405 Cas Had $M_{\rm ejecta}$$>$$M_{\rm accreted}$ and so is Unlikely to be a Type Ia Supernova Progenitor
Nova 2021 Cassiopeia (V1405 Cas) was an ordinary J(175) neon nova with the white dwarf mass estimated to be $M_{\rm WD}$=0.60$\pm$0.10 $M_{\odot}$. I found an orbital period of $P$=0.1884 days, and have tracked 20 times of photometric minima from 2013–2025. I measure that $P$ increased from before to after the eruption with $P_{\rm pre}$=0.1883919$\pm$0.0000018 days and $P_{\rm post}$=0.1884043$\pm$0.0000018 days, for $\Delta P$/$P$=66$_{-18}^{+21}$ parts-per-million. With correction for the angular momentum loss by the binary during the eruption, I derive that the nova ejected $M_{\rm ejecta}$=7.5$\times$10$^{-4}$ $M_{\odot}$, with an extreme range of (2.9–40)$\times$10$^{-4}$ $M_{\odot}$. The mass accreted during the previous eruption cycle comes from the trigger mass, and is $M_{\rm accreted}$=(1.6$\pm$0.4)$\times$10$^{-4}$ $M_{\odot}$. V1405 Cas provides counterexamples against five claims about CV evolution that have dominated since the 1980s. First, V1405 Cas has positive $\dot{P}$, and this is contrary to the Magnetic Braking Model. Second, the $\Delta P$ is 20$\times$ too small to allow the system to fade into a hibernation state. Third, $M_{\rm WD}$ is decreasing over time, as shown by $M_{\rm ejecta}$$>$$M_{\rm accreted}$ and by being a neon nova. Fourth, V1405 Cas is not a Type Ia supernova progenitor, for the same reasons. Fifth, the orbital period of V1405 Cas increased by $+$75 ppm from 2013–2025, as a counterexample to the pervasive idea that cataclysmic variables are universally declining in period from evolution. V1405 Cas is the latest of recent measures of $\Delta P$ and $\dot{P}$ for 52 cataclysmic variables and 25 X-ray binaries that have together refuted all five claims.
Read the full abstract
Information Theory
Data Protection in Function-Correcting Symbol-Pair Codes: Redundancy Bounds and Protection Profiles
In several storage systems, including DNA storage and flash memory, errors affect neighbouring symbols jointly, and the Hamming metric does not adequately capture such error patterns. The symbol-pair read channel, introduced by Cassuto and Blaum~\cite{cassuto2011codes}, addresses this by reading consecutive pairs of symbols rather than individual symbols. Motivated by this, we introduce function-correcting symbol-pair codes with data protection (FCSPC-DP), which guarantee reliable recovery of a desired function of the message while simultaneously protecting the message itself against symbol-pair errors. We derive bounds on the optimal redundancy of such codes and establish a relationship with joint-pair distance matrices. We also give explicit constructions of FCSPC-DP for locally pair-bounded functions and symbol-pair weight functions. We introduce the pair-separation constant of a function, the minimum symbol-pair distance between messages sharing a function value, and show that when it is sufficiently large, data protection requires no additional redundancy: the optimal redundancy coincides with that of the corresponding code without data protection. Considering the symbol-pair analogue of the $\alpha$-distance graph, we introduce two code invariants, the generation profile and the disconnection threshold, and use them to characterise a code's protection properties. Relating the two metrics through these invariants yields upper and lower bounds on the symbol-pair threshold in terms of its Hamming counterpart, both of which are attained. We further extend the classical Plotkin and sphere-packing bounds to this setting.
Read the full abstract
Image and Video Processing
Exponential Pixelating Integral transform with dual fractal features for enhanced chest X-ray abnormality detection
The heightened prevalence of respiratory disorders, particularly exacerbated by a significant upswing in fatalities due to the novel coronavirus, underscores the critical need for early detection and timely intervention. This imperative is paramount, possessing the potential to profoundly impact and safeguard numerous lives. Medically, chest radiography stands out as an essential and economically viable medical imaging approach for diagnosing and assessing the severity of diverse Respiratory Disorders. However, their detection in Chest X-Rays is a cumbersome task even for well-trained radiologists owing to low contrast issues, overlapping of the tissue structures, subjective variability, and the presence of noise. To address these issues, a novel analytical model termed Exponential Pixelating Integral is introduced for the automatic detection of infections in Chest X-Rays in this work. Initially, the presented Exponential Pixelating Integral enhances the pixel intensities to overcome the low-contrast issues that are then polar-transformed followed by their representation using the locally invariant Mandelbrot and Julia fractal geometries for effective distinction of structural features. The collated features labeled Exponential Pixelating Integral with dually characterized fractal features are then classified by the non-parametric multivariate adaptive regression splines to establish an ensemble model between each pair of classes for effective diagnosis of diverse diseases. Rigorous analysis of the proposed classification framework on large medical benchmarked datasets showcases its superiority over its peers by registering a higher classification accuracy and F1 scores ranging from 98.46 to 99.45% and 96.53-98.10% respectively, making it a precise and interpretable automated system for diagnosing respiratory disorders.
Read the full abstract
Combinatorics
Erdős-Sós for digraphs
It is shown that every Eulerian digraph on $n$ vertices with more than $(t-1)n$ arcs contains every oriented tree with $t$ edges. The digraphs have no loops or repeated arcs, but opposite arcs are permitted. The bound is sharp for each fixed oriented tree, as witnessed by disjoint unions of complete bidirected graphs. Previously, such tight bounds were not known, even just for directed paths. This can be considered as a directed analog of the recently proved Erdős-Sós conjecture. The result was proved by GPT-6 Astra.
Read the full abstract
Information Theory
A Mathematical Theory of Pragmatic Information
We propose a mathematical theory of pragmatic information that connects communication, control, and decision-making. Its central notion is the isoteleia mapping, which formalizes equifinality: distinct semantic paths that lead to the same optimal action are treated as pragmatically equivalent. This mapping yields a three-tier hierarchy of syntactic, semantic, and pragmatic information, in which each successive abstraction removes distinctions that are irrelevant to the task. We then define pragmatic entropy, up/down pragmatic mutual information, channel capacity, and rate-distortion, and prove lossless source coding, channel coding, and rate-distortion theorems that extend Shannon's results. These measures quantify decision uncertainty, reliable transmission, and task-oriented compression at the level of terminal actions. We further introduce pragmatic value of information (VoI) and pragmatic cost of information (CoI) as decision-theoretic duals to rate-distortion and capacity, and develop a Lagrangian dual framework for cross-layer optimization. The resulting pragmatic efficiency bound $\mathcal{E}_p(\lambda)=\sup_R[\Phi_p(R)-\lambda\mathrm{CoI}_p(R)]$ characterizes the maximum net utility attainable by a resource-constrained intelligent system under a given resource price, yielding a behavioral capacity that extends Shannon's symbol-level capacity to goal-directed action. Extensions to continuous messages provide closed-form expressions for Gaussian channels and sources, while dynamic settings are addressed through a Bellman equation for sequential decision-making. The framework supports task-oriented communication, networked control, autonomous systems, and embodied AI by shifting emphasis from symbol fidelity to the effectiveness of information in guiding actions. In this way, it offers a common language for systems that extract value from information under resource constraints.
Read the full abstract
Cosmology and Nongalactic Astrophysics
Constraining spinning primordial black holes with interstellar dust heating
Primordial black holes (PBHs) are a well-motivated dark matter candidate, and their cosmic abundance is constrained by a variety of observational probes. PBHs in the mass range $10^{15}\,\text{g}\,{-}\,10^{17}\,\text{g}$ are evaporating today via Hawking radiation, a process that can heat interstellar dust and modify its thermal emission. Recent studies have used this effect to place constraints on the abundance of non-spinning PBHs. We extend this approach by investigating the influence of PBH spin on dust-heating constraints. Furthermore, we account for secondary photons that originate not only from the decay of gauge bosons but also from the decay of hadrons produced via the fragmentation of primary quarks and gluons emitted through Hawking radiation. By comparing the dust heating rate induced by spinning PBHs with the maximum cooling rate of dust, considering both silicate and graphite grains, we derive new upper limits on the fraction of dark matter in the form of PBHs, $f_{\rm PBH}$. Our results show that the constraints depend on both PBH mass and spin. Smaller PBHs with higher spin yield stronger limits. For example, in the cases we investigated, the strongest constraint is $f_{\rm PBH} \sim 1.5 \times 10^{-4}$ for $M_{\rm PBH} = 10^{15}{\rm g}$ and spin parameter $a_{*} = 0.9999$. Although these limits are less stringent than existing constraints in the same mass range, they provide a distinct and complementary approach to constraining the abundance of PBHs.
Read the full abstract
Quantum Physics
Full Inseparability and Genuine Multipartite Entanglement Coincide for Finite-Mode Gaussian States
For general mixed states, entanglement across every bipartition need not imply genuine multipartite entanglement (GME), because a biseparable decomposition may switch the separable cut from term to term. We prove that this convex ambiguity disappears for Gaussian states of finitely many bosonic modes. More generally, for any finite family of partitions, a Gaussian density operator in the trace-norm-closed convex class generated by states separable across those partitions is already separable across one fixed partition in the family. Only the target is Gaussian; a valid decomposition may be continuous and may contain arbitrary non-Gaussian states. Thus full inseparability and GME coincide, Gaussian k-separability and k-producibility reduce to fixed-partition tests, and party-wise tensor powers cannot activate GME from a biseparable Gaussian state. The proof combines a spectral selector with a holomorphic rigidity argument that converts one product vector in the square-root range of a Gaussian state into a block-local covariance certificate. The result shows that partition mixing, a generic mixed-state mechanism, adds no new exact finite-mode Gaussian states.
Read the full abstract
Numerical Analysis
A variational physics-informed graph neural network for heterogeneous solid mechanics
Stress localization in heterogeneous solids is governed by the bimaterial interface, where the displacement field remains $C^0$-continuous, while in-plane stresses jump due to the stiffness mismatch. Coordinate-based physics-informed neural networks (PINNs) represent this jump via a prescribed regularization width or a weighted interface penalty, making their accuracy sensitive to how phase-contrast changes are handled. This work presents a variational, label-free physics-informed graph neural network (PI-GNN) in which the heterogeneity is carried by the discretization rather than by the trial field. The solver operates on a conforming adaptive mesh graph, assigns constitutive behavior per element, and minimizes the discrete total potential energy as a single unweighted objective in which only first derivatives appear. The discrete energy on piecewise-linear elements coincides with the finite element (FE) Ritz functional. Dirichlet conditions are enforced by construction, with no penalty term, no interface weight, and no prescribed transition width. Using one fixed architecture, optimizer, and loss across small-strain elasticity and finite-strain Neo-Hookean hyperelasticity in two and three dimensions, the von Mises error remains below $3.58\%$ across a stiffness-contrast sweep spanning $(E_{\mathrm{inc}}/E_{\mathrm{mat}}\in[10^{-2},10^{2}])$, where a strong-form PINN degrades to $5.58\%$, and its displacement error reaches $7.66\%$ against $0.49\%$ for the PI-GNN. A trained network halves the ($\sigma_{xx}$) error of an energy-based PINN ($5.01\%$ versus $10.94\%$). Training cost exceeds a single FE solve by more than an order of magnitude, so the construction is a variationally consistent, penalty-free interface representation for parametric surrogates and inverse identification rather than a replacement for a one-off FE analysis.
Read the full abstract
Graphics
ReCHOIR: Contact-guided Human Object Interaction Retargeting to Diverse Characters
We present ReCHOIR, a novel contact-guided motion retargeting method for transferring human object interaction (HOI) motions across diverse humanoid characters. Unlike prior motion retargeting methods that primarily focus on transferring human motion alone, our goal is to preserve not only the semantics of the original body movement but also consistent interaction between the character and the manipulated object, while jointly producing aligned target human and object motions. Given source HOI motion, object geometry, and contact cues extracted from the source interaction, ReCHOIR retargets an HOI sequence to target characters with different skeletal configurations while maintaining both motion semantics and contact-consistent interaction patterns. Our method builds on a Part-Aware Motion Embedding (PAME) autoencoder, which encodes full-body motion into a shared body-part-wise latent space. This representation enables generalization across heterogeneous skeletons while preserving local motion semantics beneficial for part-aware adaptation in HOI retargeting. On top of this representation, we introduce a contact-guided retargeting module and an object motion decoder for HOI retargeting. The contact-guided retargeting module treats the source object interaction as a condition for refining target character motion: object- and contact-related signals are encoded into a body-part-aligned latent representation and injected into decoding through a residual control branch, enabling stronger adaptation in interaction-relevant body regions without discarding the underlying motion prior. In parallel, the object motion decoder predicts a target object motion aligned with the refined target character motion, ensuring that the object trajectory remains consistent with how the interaction is realized by the target character.
Read the full abstract
Machine Learning
Thompson Sampling for Non-Monotone Convex Ridge Bandits: Monotonicity Is Not Needed for Polynomial Regret
Bakhtiari, Lattimore and Szepesvári (COLT 2025) proved that Thompson sampling (TS) has Bayesian regret $\tilde O(d^{5/2}\sqrt n)$ for bandit convex optimisation with convex monotone ridge losses $f(x)=\ell(\ip{x}{\theta})$, and asked whether monotonicity of the link is necessary. We give a qualitative negative answer. For every prior on $[0,1]$-valued, $1$-Lipschitz convex ridge losses with an arbitrary convex, possibly non-monotone, link, and for any fixed measurable selection of minimisers, exact-posterior TS has Bayesian regret $O\big((d+1)^4\sqrt{dn}\,\log(e+nd\max\{1,\diam K\})\big)=\tilde O(d^{9/2}\sqrt n)$. The monotone proof relies on a single-removal John-ellipsoid dichotomy; we show by an explicit twelve-point configuration that this dichotomy fails for non-monotone links, and replace it by an $O(d^2)$ cardinality bound for “uninformative” configurations. The bound uses a Boolean rounding argument: a $0$-$1$ matrix within $1/(4r)$ in max-norm of a rank-$r$ matrix has rank at most $2r-1$. We construct $d(d+1)$ uninformative losses, showing that the cardinality bound is tight up to constants in the large-diameter-to-gap regime, and give a self-contained information-ratio-to-regret transfer that is uniform over fixed measurable selections. Whether the $d^{5/2}$ dependence of the monotone case can be retained remains open.
Read the full abstract
Machine Learning
EGGROLL, Unrolled: Understanding and Improving Low-Rank Evolution Strategies at Scale
EGGROLL (Sarkar et al., 2026) makes evolution strategies (ES) practical for LLMs by replacing dense Gaussian weight perturbations with low-rank Gaussian products, often of rank one. This choice is computationally attractive but geometrically severe: Each rank-one perturbation lies in a zero-volume subset of the ambient matrix space, despite having identity covariance. We characterize the EGGROLL update mean field at finite rank and nonzero perturbation radius as a resolvent applied to the gradient of the perturbation-smoothed objective. This transformation can make the mean field nonconservative and reverse the local stability of an optimum. EGGROLL nevertheless recovers the gradient exactly on quadratic objectives at every rank and radius. In finite populations, the additional sampling variance of rank-one perturbations relative to dense Gaussian ES decays inversely with matrix width under a local affine model, and is only $0.098\%$ at width $4096$. Finally, we introduce LOO-ROLL, a leave-one-out estimator that replaces EGGROLL's two antithetic evaluations per direction by one. At equal evaluation cost, LOO-ROLL halves estimator MSE in transformer blocks. Across fourteen post-training settings up to 14B parameters, matched-time comparisons with EGGROLL yield eleven improvements in individual paired tests and no significant loss. At 1.7B, 8B, and 14B parameters, matched-time gains are $2.9$, $14.1$, and $7.9$ percentage points on GSM8K and $12.2$, $8.4$, and $8.1$ points on MATH-500.
Read the full abstract
Materials Science
Silicon Solar Cell Design for >30% Efficiency via Singlet Fission
Singlet fission (SF) materials convert high-energy photons into multiple charge carriers, providing a route to exceed the efficiency limits of single-junction silicon solar cells without many of the complexities of multi-junction tandem designs. Following the first demonstration of an SF-enhanced silicon solar cell in 2025, there is a need to understand how SF materials can be effectively integrated into high-efficiency industrial silicon devices and translated from proof of concept to a manufacturable technology. Using coupled optical and electrical simulations, we assess the efficiency potential of several industrially relevant silicon cell architectures combined with SF materials. Interdigitated back-contact (IBC) cells offer the greatest potential for improvement due to unrestricted front-surface access and can achieve efficiencies exceeding 33%. However, performance is highly sensitive to front-surface passivation quality. Appropriate silicon design, particularly controlled surface doping and fixed interfacial charge, can mitigate recombination losses and relax passivation requirements for ultra-thin exciton-transfer layers.
Read the full abstract
Combinatorics
Fractional DP-colorings of $d$-degenerate locally sparse graphs
Bernshteyn, Kostochka, and Zhu (2020) introduced the notion of fractional DP-coloring, which generalizes both fractional coloring and fractional list coloring. Among several foundational results, they proved that every $d$-degenerate bipartite graph $G$ satisfies $\chi_f^{\mathrm{DP}} \le (1 + o(1))\frac{d}{\log d}$, and that this bound is optimal—a stark contrast to ordinary fractional coloring. In this paper, we extend this upper bound to all $d$-degenerate triangle-free graphs, proving that $\chi_f^{\mathrm{DP}} \le (4 + o(1))\frac{d}{\log d}$. This generalizes a recent result of Martinsson and Steiner (2025) for ordinary fractional coloring. We derive this result as a corollary of a more general upper bound concerning locally sparse graph orderings. Specifically, a $d$-degenerate graph $G$ is left $k$-locally-sparse if it admits a degeneracy ordering in which, for every vertex $v$, the subgraph induced by its back-neighbors contains at most $k$ edges. We show that if a $d$-degenerate graph $G$ is left $\frac{d^2}{f}$-locally-sparse, then \[ \chi_f^{\mathrm{DP}}(G) \le (8 + o(1))\frac{d}{\log f}. \] This immediately yields an identical upper bound on the ordinary fractional chromatic number $\chi_f(G)$, improving upon the leading constants of previously known bounds. Additionally, we establish the asymptotic sharpness of this result up to the leading constant. For any $1 \ll f \le d^2$, we construct $d$-degenerate graphs that are left $\frac{d^2}{f}$-locally-sparse and satisfy $\chi_f(G) \ge (1 - o(1))\frac{d}{\log f}$. Finally, as applications of our main theorem, we obtain improved upper bounds on the fractional DP-chromatic number of $d$-degenerate $K_{1,t,t}$-free graphs, as well as $K_{t,t,t}$-free graphs with maximum degree $\Delta$. Notably, these bounds improve upon existing results even in the setting of ordinary fractional coloring.
Read the full abstract
Materials Science
A model of grain growth in UN integrating molecular dynamics, phase-field modeling, and uncertainty quantification
Grain growth kinetics and grain-boundary (GB) properties in uranium mononitride (UN) are investigated through an integrated multiscale framework combining molecular dynamics (MD), phase-field modeling, and surrogate-assisted uncertainty quantification. MD simulations yield GB energies for 27 symmetric tilt boundaries from 0–2000~K, which are consistent with available DFT values. The average GB energy is nearly temperature-independent below 1000~K and increases at higher temperatures. A mechanistic pore-drag model applied to the only available grain growth dataset for actinide nitrides yields a mobility reduction factor of $s \approx 0.93$–$0.99$, statistically indistinguishable from unity, confirming that pore drag is negligible under the experimental conditions. The intrinsic GB mobility is therefore extracted directly from the effective mobility, yielding $M_0 = 2.05\times10^{-15}$~m$^4$/(J$\cdot$s) and $Q_M = 0.89$~eV. Phase-field simulations conducted from 1500–2000~K confirm normal curvature-driven grain growth, with grain size distributions converging to the Hillert-like form. A surrogate-assisted global sensitivity analysis—combining principal component analysis, Gaussian process regression, and Sobol decomposition—reveals that the mobility prefactor $M_0$ dominates output variance at all times, followed by the activation energy $Q_M$, while the GB energy $\gamma$ contributes minimally. These results establish the first quantitative grain growth framework for UN and identify the reduction of uncertainty in $M_0$ and $Q_M$ as the highest-priority target for future experimental efforts.
Read the full abstract
Machine Learning
Phases in a class of associative memories via hidden neurons
Associative memory in the Hopfield network is attractor dynamics in a disordered many-body system, and higher-order and exponential extensions turn its retrieval update into softmax attention. The polynomial and exponential regimes have been analyzed by different methods, with no common architecture in which to ask what fixes the storage scale. In this paper we study the bipartite architecture of Krotov and Hopfield, which we call the class $H$, whose model is fixed by a Lagrangian for each layer, taking the hidden neurons as the order parameter of retrieval. At polynomial load the replica method yields the replica-symmetric phase diagrams and closed-form capacities, and the crosstalk moment is common to Ising and spherical visible neurons, so their differences come from the visible entropy. With a softmax hidden layer the load is exponential, and a copy representation maps the thermodynamics onto random-energy-model counting, with paramagnetic, condensed, and frozen phases. Heating destabilizes retrieval by quantized reassignments of attention, and typical Gaussian patterns remain metastable at every load. The regimes differ in their crosstalk statistics, central-limit at polynomial load and large-deviation at exponential load, and the class $H$ splits retrieval into two roles, the visible Lagrangian fixing stability and the hidden one the storage scale, two axes that may also guide the design of new Lagrangians.
Read the full abstract
Combinatorics
Balanced even cycles in signed graphs:Turán bounds, double covers, and parity obstructions
We study Turán problems for balanced even cycles in simple signed graphs, where signed subgraphs are considered up to switching. For every balanced bipartite signed graph, the signed and ordinary Turán numbers differ by at most a factor of two. Our main structural results concern the underlying graphs that admit a signing in which every $2k$-cycle is unbalanced. We characterize these graphs by the absence of an odd dependence among their $2k$-cycle incidence vectors, give a cohomological formulation, and construct subgraph-minimal obstructions of arbitrarily large order. In particular, there is no finite forbidden-subgraph characterization. We also give an exact closed-walk criterion for cycles in double covers and derive a direct signed breadth-first-search upper bound. As applications, we prove \[ \hex(n,C_{+4})=\left(\frac{\sqrt2}{2}+o(1)\right)n^{3/2} \] and study the signed hexagon number $R_6(n)=\hex(n,\{C_{-3},C_{+6}\})$. We characterize the underlying graphs counted by $R_6$ and express it as an extremal problem for ordinary $C_6$-free graphs with a prescribed involution. For every sufficiently large $n$, we construct examples with $\Omega(n^{4/3})$ edges, and we give an equivariant construction attaining the coefficient obtained from the Füredi–Naor–Verstraëte lower bound by double-cover transfer. Finally, we give $n$-vertex $C_{+10}$-free signed graphs with $\Omega(n^{6/5})$ edges and use octagon examples to illustrate the limitations of theta-freeness as a signing criterion.
Read the full abstract
Quantum Physics
Sample-optimal learning of stabilizer states
It is well-known that learning a pure $n$-qubit stabilizer state $|\psi\rangle$ both requires, and can be accomplished with, access to a number of copies of $|\psi\rangle$ linear in $n$. However, the precise constant coefficient of this scaling does not appear to have been determined. Here we prove that $L_\delta(n)$, the smallest number of copies from which a quantum procedure can identify any stabilizer state with failure probability at most $0<\delta<1/8$, satisfies $n+\lceil\log_2(1/\delta)\rceil-3\leq L_\delta(n)\leq n+\left\lceil\log_2(1/\delta)\right\rceil+4$. We present a polynomial-time quantum learning algorithm that saturates this bound, achieving a constant factor improvement in sample-complexity over previously known approaches. As an immediate corollary, we obtain via the Choi-Jamiolkowski isomorphism an algorithm for learning an unknown $n$-qubit Clifford unitary from $2n+\left\lceil\log_2(1/\delta)\right\rceil+4$ queries, the $n$-dependence of which we show to be optimal. Our proof technique, which involves Fourier analysis on the abelian group $\mathbb{Z}_4^n \times \mathbb{F}_2^{n(n-1)/2}$, seems to be qualitatively different to previous approaches to stabilizer state learning, and may be of some independent interest; in particular, it admits natural generalisations to further problems in quantum learning theory.
Read the full abstract
Materials Science
Measuring chiral phonons
Chiral phonons are quantized vibrations where the atomic motion in a solid breaks improper rotation symmetries. In many cases, chiral phonons possess angular momenta and are therefore selective to circularly polarized light. Both fundamental and applied research efforts on chiral phonons have been gaining increasing attention owing to their importance in a variety of fields including spintronics, spin-selective chemical reactions, thermal transport, quantum information processing and biosensing, where the bi-directional spin-lattice coupling enabled by chiral phonons can be harnessed in new ways, and potentially lead to new functionalities. Thus far, the studies of chiral phonons across diverse materials platforms have evolved largely independently within these fields, but the experimental techniques are often interrelated. In this perspective, we present a detailed description, as well as advantages and disadvantages of the current approaches for experimentally measuring chiral phonons in chiral and achiral materials. We conclude with a discussion of new methods for measuring chiral phonons. Ultimately, this work seeks to offer an experimental guide for systematically investigating the properties of chiral phonons in various materials systems and applications.
Read the full abstract
High Energy Physics (Theory)
Landau-Ginzburg description of an exceptional ${\mathcal N}=1$ minimal model
The $\mathcal N=1$ superconformal minimal model with $m=12$ and the exceptional modular invariant $(E_6,D_8)$ is the unitary minimal model of the super-$W_3$ algebra. We propose its Landau-Ginzburg description using two real scalar superfields with the cubic superpotential ${\cal W}=g_1 XY^2/2 + g_2X^3/6$. For $g_1=g_2$, this superpotential is known to describe a product of two $m=3$ $\mathcal N=1$ superconformal minimal models, which is the $m=10$ model with the $(D_6,E_6)$ modular invariant. The exceptional $m=12$ superconformal minimal model is realized at a different fixed point of the same theory. Testing this Landau-Ginzburg description requires the fusion ring of the minimal model, which we obtain from the modular data of the extended algebra. The fusion ring has a $\mathbb Z_2$ grading by chiral fermion parity that the ordinary fusion coefficients do not determine. This grading, composed with conjugation, gives the generator of the R-parity $\mathbb Z_2^{R}$ of the Landau-Ginzburg theory. We then treat the theory with superpotential $\cal W$ as a Gross-Neveu-Yukawa model in $d=4-\epsilon$ and find a weakly coupled infrared fixed point with $g_1/g_2=3/2+\mathcal O(\epsilon)$, at which supersymmetry emerges. We also describe the renormalization group flow from this fixed point to the decoupled fixed point with $g_1=g_2$. The operator dimensions at the coupled fixed point, continued to $d=2$, agree approximately with their values in the $m=12$ superconformal minimal model. Finally, we estimate the scaling dimensions in the new interacting $d=3$ $\mathcal N=1$ superconformal field theory.
Read the full abstract
Econometrics
Experimental Design for Policy Choice
We show how to optimally design experiments when the resulting data will be used to choose a welfare-maximizing policy subject to constraints. A decision maker seeks to maximize Bayes expected welfare by choosing a policy whose effects depend on an unknown finite-dimensional parameter. The decision maker has access to a first wave of experimental data with a fixed design but may choose the design of a second wave that will be collected before choosing the policy. The resulting experimental design–policy choice problem is a very high-dimensional dynamic program that is generally intractable in finite samples. We propose a tractable approximation based on the limit experiment and show it is asymptotically optimal using a new asymptotic representation theorem for adaptive experiments with continuous treatments. We apply the method to a conditional cash transfer experiment and demonstrate the potential for large gains from tailoring the experiment to the policy choice.
Read the full abstract
Hardware Architecture
Fengshui: Demystifying Chiplet Ecosystem and Bespoke Neural Network Accelerator Codesign
Modern ML workloads, with stringent latency and energy constraints, are increasingly hard to run efficiently on homogeneous commodity hardware. We argue that operator-level disaggregation–tailoring microarchitecture, batching, and memory hierarchy to each operator–is essential to overcome these limitations, though the resulting highly bespoke accelerators incur prohibitive Non-Recurring Engineering (NRE) costs. Chiplet-based integration amortizes NRE across applications, but choosing which chiplets to build and how to compose them into accelerators is circularly dependent–a chiplet pool's value depends on the constructed accelerators, while accelerator quality is constrained by available chiplets. This paper introduces Fengshui, a chiplet ecosystem and accelerator co-design framework that jointly optimizes chiplet pool composition and bespoke application-specific integrated circuit (BASIC) design. Fengshui constructs BASICs through operator-level disaggregation, co-exploring chiplet and memory heterogeneity, tensor fusion, and pipeline/tensor/expert parallelism with place-and-route validation for physical implementability. With just 8 strategically selected chiplets, encompassing network switches, processing-in-memory units, and accelerators with diverse microarchitectures, Fengshui-generated BASICs achieve 48.5%, 88.1%, 93.0%, and 97.8% reductions in energy, energy-cost product (EC), energy-delay product (EDP), and energy-delay-cost product (EDPC) over homogeneous accelerators, while scoring within 4.1% of unconstrained heterogeneous designs across diverse neural networks. For datacenter MoE and dense LLM serving, Fengshui reduces prefill energy and EC by up to 16.8% and 28.7%, respectively; for edge autonomous vehicle perception, it achieves 12.0% energy and 23.6% EC reductions under real-time latency constraints.