Peer Revenue Equity in science

Papers

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

All of arXiv

Friday 6 November

Artificial Intelligence

Learning Whom to Trust About the Unseen: Source Credibility and Reliability Transfer in Language Models

A source who has been right before deserves attention, but about what? We ask how far a demonstrated record of reliability travels in language models when the new claim cannot be checked, and whether the answer depends on the kind of claim. Three behavioral studies across six models approach the question from three angles. Study 1 adapts the Einstein-effect paradigm and finds that models, unlike the human sample, favor a spiritual leader over a scientist for the same obscure statement, and that the preference largely reverses when the statement is reworded in the vocabulary of physics. Study 2 adapts a paradigm in which a trusted expert calls a scientific or religious explanation unknown or a mystery; models reproduce the human domain asymmetry in exaggerated form. Study 3, the central test, gives two fictional informants different records on checkable river forecasts and then has them endorse matched claims about hidden natural and spiritual causes of the same village events. Most models carry the record to natural causes and barely at all to spiritual ones; one model transfers it weakly to either. Telling the model that the reliable informant forecasts by spiritual guidance rather than by measurement lowers that informant's influence on natural claims in every model and produces no comparable gain on spiritual claims in any. Demonstrated reliability thus has a scope in these models, and the account a source gives of how they came to be reliable can narrow it. Whether this reflects a judgment about relevance or the low prior credence of spiritual claims is the question the design leaves open, and we outline the experiment that would settle it.
Read the full abstract

Human-Computer Interaction

"I had nobody to turn to": Why LGBTQ+ People Choose (Not) to Use AI Chatbots for Personal Health and Relationship Advice

While there is an abundance of research into large language model (LLM) biases against LGBTQ+ communities, far less research has examined how and why LGBTQ+ people choose to make use of AI chatbots. In this work, we examine how LGBTQ+ people choose (not) to use AI chatbots across three domains in which LGBTQ+ users have unique needs: relationship support, mental health support, and gender or sexual health support. We surveyed 73 people (60 LGBTQ+; 13 non-LGBTQ+) on how they interact with AI chatbots and find that participants' access to external resources, privacy concerns, and perceptions of chatbot capabilities shape whether and how they choose to make use of AI chatbots. For instance, some LGBTQ+ participants felt that chatbots might actually be less biased against LGBTQ+ people than the people they know offline. We also find that LGBTQ+ participants report using AI chatbots less frequently than non-LGBTQ+ participants across 15 specific personal health and relationship advice scenarios, citing concerns about AI sycophancy and adverse cognitive side-effects. Recognizing AI chatbots as vital (albeit imperfect) resources for LGBTQ+ people, we recommend researchers consider systemic issues that drive users to AI chatbots and design technologies (AI or otherwise) that prioritize privacy, anonymity, and real human connection.
Read the full abstract

All authors Jacob Hobbs, Ashray Gupta, Dennis Ulmer, Geetanjali Aich, Nolan Golden, Yanan Long, Ian Rios-Sialer, Jennifer Mickel, Morgan Klaus Scheuerman, Rohit Avinash Mujumdar, Virginia K. Felkner, Jordan Taylor, William Agnew

Genomics

Synchronized Optimal Transport Reconstructs Coupled Transcriptomic and Chromatin Dynamics

Single-cell RNA and ATAC data collected across developmental stages provide time-course snapshots of transcriptional and regulatory states. Since sequencing is destructive, trajectories between these snapshots must be inferred from independently sampled cell populations. Existing approaches typically infer continuous dynamics in a single modality or integrate RNA and ATAC in a shared representation. Here we present Cross-Omics Aligned Trajectory Inference (COATI), which extends unbalanced synchronized dynamical optimal transport to time-dependent cross-modal maps for reconstructing coupled trajectories in distinct RNA and ATAC state spaces. We evaluated COATI on synthetic benchmarks and four developmental datasets using held-out-stage reconstruction and lineage analyses. COATI improved reconstruction of cell populations at held-out stages, including recovery of chromatin-state heterogeneity within cell types, and better preserved developmental lineage relationships. The unbalanced formulation allows the inference of population mass changes, which were shown to associate with DNA-replication programs at early stages and lineage-maturation programs later in development. The coupled trajectories enabled analysis of lineage-specific transcriptional and chromatin changes and their temporal relationships during development. Lineage responses inferred by COATI agreed with independent CRISPR perturbation results, and the inferred regulatory patterns were consistent with independent epigenetic data.
Read the full abstract

All authors Jingfeng Wang, Yanxiang Zhao, Qing Nie, Zixuan Cang

Representation Theory

Plus Constructions and Canonical Induction for Green Categories

We develop canonical induction for Green categories associated with Mackey systems. We establish a change-of-rings framework in which the lower-plus and upper-plus constructions are realized by induction and coinduction of modules. The two plus adjunctions yield the mark morphism, and, when the relevant group orders are invertible in the coefficient ring, Möbius inversion gives an explicit inverse. Using the resulting linearization and tom Dieck maps, we construct a canonical induction map, give an explicit formula for it, and characterize when it is a right inverse of linearization. In the classical setting of Mackey functors, our construction and splitting criterion recover Boltje's results.
Read the full abstract

Materials Science

Conformationally frustrated epitaxy at heteromolecular interface

Precise control of molecular heteroepitaxy offers a route to multilayer architectures with well-defined interfaces. However, the role of intramolecular flexibility in epitaxial matching remains insufficiently understood. Here, we investigate the heteroepitaxy of intact 1,3,5-tris(4-carboxyphenyl)benzene (BTB) on a molecular $\delta$-layer composed of fully deprotonated BTB molecules on Ag(111). Combining scanning tunneling microscopy, low-energy electron microscopy and diffraction, X-ray photoelectron spectroscopy, and density functional theory, we identify the key interactions controlling molecular packing and epitaxial growth. We show that intermolecular bonding motifs are preserved on the molecular template, but their lattice parameters and molecular conformations change as compared to a silver surface. Individual molecules exhibit distinct appearances in scanning tunneling microscopy, consistent with site-dependent conformational variations identified by theory. Our calculations reveal that the competition between molecule-substrate interactions and conformational energy produces different compromise geometries, resulting in conformational frustration at the epitaxial interface. Crucially, constraining BTB into a planar geometry changes the preferred lateral adsorption positions, demonstrating that molecular flexibility directly modifies the energetic preferences for epitaxial registry. Comparison with more rigid molecular overlayers further supports this interpretation. Our findings reveal that molecular flexibility contributes to epitaxial matching by allowing unfavorable molecule-substrate interactions to be compensated by energetically favorable conformations while preserving the intermolecular network.
Read the full abstract

All authors Jakub Planer, Pavel Procházka, Tomáš Krajňák, Veronika Stará, Daniela Hrubá, Zdeněk Endstrasser, Tomáš Skála, Jan Čechal

Signal Processing

Stochastic High-Resolution Parameter-Based Channel Modeling for Sub-Terahertz Outdoor Device-to-Device Propagation at 145 GHz

Since device-to-device (D2D) systems operating in the sub-THz regime are of interest for a variety of 6G applications, modeling their propagation channels is therefore essential. This paper presents high-resolution parameter estimation (HRPE) of specular multipath components (MPCs) as well as diffuse multipath components (DMCs) from an extensive measurement campaign. The extracted MPCs and DMCs are then clustered, and statistical models for both intra- and inter-cluster statistics are established. Among other insights, we find a large number (typically 10-50) of clusters; both the powers and densities of MPCs within each cluster show significant variations. DMC occurs both when associated with discrete MPCs and standalone. The paper provides a complete model that can generate double-directional impulse responses, and a MATLAB program implementing it is made available. Extensive verifications of "derived" (not directly modeled) parameters are provided as well.
Read the full abstract

All authors Zihang Cheng, Naveed A. Abbasi, Jorge Gomez-Ponce, Jianzhong Zhang, Andreas F. Molisch

Human-Computer Interaction

Lea: An Agent Backbone for Mathematician-Controlled Formalization in Lean

Agentic systems now formalize research-level mathematics in Lean, but they are built for autonomy: the mathematician states a target and inspects a result. The alternative, a general-purpose coding harness, is domain-neutral and must be assembled into a formalization workspace before any mathematics begins. Mathematicians, however, report wanting to keep high-level control over how an argument is decomposed and established. We present $\textbf{Lea}$, a Lean-specific agent backbone for mathematician-facing software rather than a single interface. Lea exposes runs, sessions, and typed events through one application-neutral API, and ships with Lean tooling (blueprints, LSP, Loogle, SafeVerify) already wired. We report two applications built on it: $\textbf{LeaChat}$, a web client for interactive proof authoring, and $\textbf{LeaOverleaf}$, which formalizes theorems annotated directly in an Overleaf document; a session begun in either continues in the other. In a case study, a mathematician outlined a proof of the Durfee square generating function in a few natural-language messages, and Lea produced a complete Lean development from Mathlib alone, following the intended decomposition closely and requiring substantive correction at a single point. Lea and both applications are released as open source so that further mathematician-facing tools can be built on the same backbone.
Read the full abstract

All authors Shaswat Patel, Daniel Aarao Reis Arturi, Kevin Gomez, Jaume de Dios Pont, Jorge Poco, Eduardo V. Teixeira, Sam Westrick, Claudio Silva, Chinmay Hegde

Optics

Self-injection-locked dark-soliton Kerr optical frequency combs with high even-order normal dispersion

We numerically investigate dissipative dark-soliton (DDS) microcombs under the action of pure high even-order normal (HEON) dispersion in a microresonator with the self-injection locking (SIL). Pure 4th, 6th, 10th, and 14th-order normal dispersion can support DDSs formed by interlocked switching waves, directly excited through a SIL-based turnkey scheme without detuning sweeps. Compared to the pure 4th-order case, the switching-wave tail oscillation period increases and its decay rate decreases with the increase of the dispersion order. Although the overall comb bandwidth narrows, the number of non-pump comb lines above a common absolute power threshold increases. Furthermore, the feedback phase controls the DDS duty cycle and intracavity comb-line power distribution, while the increase of the pump power improves the estimated pump-to-comb output conversion efficiency.
Read the full abstract

All authors Lei Shi, Rujiang Li, Shilong Liu, Dumitru Mihalache, Boris A. Malomed, Pengfei Li

High Energy Astrophysical Phenomena

Unlocking Pandora's Secrets: Probing the Thermodynamics of Abell 2744

Abell 2744 (Pandora's Cluster) is a massive galaxy cluster undergoing multiple mergers, characterized by a highly disturbed X-ray morphology, elevated intracluster medium (ICM) temperatures, and several radio relics associated with merger driven shocks. We present NuSTAR observations of Abell 2744 aimed at mapping its thermodynamic structure and constraining the presence of inverse Compton (IC) scattering from relativistic electrons. We measure the temperature distribution across the cluster and identify a prominent shock front with a Mach number of $\mathcal{M} = 1.58^{+0.42}_{-0.30}$, aligning with the southeastern boundary of the radio halo. The global NuSTAR spectra are modeled with single- and two-temperature thermal components, as well as thermal plus power-law models to test for IC emission. The spectra are well described by purely thermal models with an average temperature of $kT \approx 11$ keV, and we find no significant evidence for non-thermal hard X-ray emission. We derive a 90% confidence upper limit on the $20-80$ keV IC flux of $ F_{\rm nt} < 2.31 \times 10^{-12} erg~s^{-1}~cm^{-2}$. Combining this with 1.5 GHz diffuse radio measurements from the VLA, we estimate a lower limit on the average magnetic field strength of $B > 0.2 \ \rm{\mu G}$.
Read the full abstract

All authors Randall Rojas Bolivar, Akos Bogdan, Daniel Wik, Gerrit Schellenberger

Artificial Intelligence

When F1 Is Not Enough: Auditing Codebooks for LLM-Assisted Annotation

Large language models (LLMs) are increasingly used for text annotation in social science, but standard performance metrics do not explain why errors occur. Low F1 may reflect limitations of the annotation system, defects in the written codebook, or inconsistency in reference labels. We propose a pre-deployment codebook audit for LLM-assisted annotation. The audit uses class-specific metrics and confusion matrices to identify problematic classes and boundaries, structured document-level review to attribute error sources, and targeted codebook revision followed by held-out evaluation. We apply the audit to two published multi-class codebooks from distinct domains. In both applications, the audit identifies under-specified or conflicting coding rules, and targeted revisions reduce the diagnosed errors on held-out data. Analyses with two LLM annotators show that some problematic boundaries and codebook revisions transfer across models, whereas other revision effects are annotator-specific. The results show that LLM annotation errors can help identify weaknesses in the written measurement instrument, not only in the model. Codebook auditing complements prompting, retrieval, and fine-tuning by clarifying whether persistent errors call for model improvement, codebook revision, or renewed scrutiny of reference labels.
Read the full abstract

Cryptography and Security

Jev as an LLM Safety Guardrail: An Empirical Study of Input Filtering, Selective Output Review, and Over-Refusal

External safety filters can prevent unsafe requests from reaching a large language model (LLM), but may also reject legitimate queries. We present an end-to-end empirical evaluation of Jev, a probabilistic decision model, as a safety guardrail for a Qwen2.5-1.5B-based pipeline. Three configurations are compared on 970 benchmark prompts: an unfiltered LLM, Jev input filtering, and tiered Jev input filtering with selective output review. Across 720 prompts labeled harmful (520 AdvBench and 200 XSTest-unsafe), the fraction recorded as answered decreases from 527/720 (73.2%) without filtering to 17/720 (2.4%) with input filtering. On 250 XSTest-safe prompts, however, non-answer/refusal increases from 3/250 (1.2%) to 27/250 (10.8%). Tiered filtering yields 40/720 (5.6%) harmful-prompt answers and 10/250 (4.0%) benign-prompt refusals, with 79/970 (8.1%) of prompts requiring output review. Joint optimization of the tiered thresholds improves macro-F1 from 0.9351 to 0.9512, while reducing harmful-prompt answers and output-review frequency at the same benign-refusal rate. Matched-budget comparisons reveal modest additional gains over a retuned single gate on the evaluation data, although these gains are not consistently observed on a held-out split.
Read the full abstract

Quantum Physics

From spectroscopy to quantum dynamics: assessing reduced models of a NaCs molecular gate

We investigate how spectroscopic calibration and basis selection affect predictions for a dipolar-exchange gate between two optically trapped $^{23}$Na$^{133}$Cs molecules. We construct a 128-state molecular Hamiltonian restricted to the $N=0,1$ rotational manifold, fit two effective optical parameters to three bright microwave transitions, and compare the resulting spectrum with four additional lines. In the coherent pair model, changing the optical polarization angle by $+0.1^\circ$ reduces the calculated gate fidelity from 94.64% to 1.09% at fixed carrier frequencies. Retuning the carriers and restoring the target Rabi frequencies recovers 89.31%. The remaining sensitivity reflects changes in the multilevel pair dynamics that are not removed by target-frequency and target-Rabi recalibration. Enlarging the molecular basis produces nonmonotonic changes in fidelity, although the 11-state and 25-state calculations give closely agreeing logical maps after phase alignment. Two controls with isolated transfer probabilities near 98.45% yield complete-gate fidelities differing by more than 11 percentage points. We relate these comparisons to a sufficient bound on the error of a projected propagator. The numerical tests quantify agreement among selected bases for a specified coherent Hamiltonian.
Read the full abstract

Statistical Methodology

Bayesian Unit-level Quantile Modeling with Application to Wealth Data

Modeling wealth quantiles over time and space and across fine-grained demographic classifications is of vital importance to policy makers. However, such data present unique challenges for standard small area estimation methodologies because they are highly skewed and heavy-tailed with both positive and negative values. Traditional area-level modeling and design-based approaches struggle to capture these features. Meanwhile, existing unit-level methodology frequently treats the survey design as ignorable and does not apply to surveys that are repeated over time. Often the analysis is restricted to alternative response variables that are non-negative, such as poverty indicators or income. To overcome these limitations, we propose a Bayesian, asymmetric Laplace pseudolikelihood model that is able to capture the major features of wealth while adjusting for survey weights and accounting for both spatial and temporal dependence. We present a Gibbs sampling algorithm for our model, as well as a variational Bayes algorithm for fast computation. We apply the method to the Survey of Income and Program Participation to characterize geographic, temporal, and demographic variation in the distribution of U.S. household wealth, including for sparsely sampled domains where direct estimates are unavailable or unstable.
Read the full abstract

Computational Engineering

DrivAerRim: A large-scale full-vehicle CFD dataset for wheel rim aerodynamics

The wheel region contributes substantially to vehicle aerodynamic drag, and rim geometry strongly influences the flow in this region. However, publicly available automotive aerodynamics datasets offer limited variation in rim design. We present DrivAerRim, a large-scale dataset of full-vehicle computational fluid dynamics simulations covering 904 DeepWheel rim designs with diverse spoke layouts, opening areas and topologies. All cases use the same DrivAer Estateback, operating conditions and numerical setup, with rim geometry as the sole design variable. Wheel rotation is modelled using moving reference frame regions around the rims and rotating wall conditions on the deformed tyres. The numerical setup is assessed through mesh sensitivity analysis and comparisons with wind tunnel measurements. The resulting 9.32 TB release includes wheel geometries, vehicle and component force coefficients, accumulated drag, surface pressure, wall shear stress, and flow fields around the vehicle and wheels. Together, these data support surrogate model training and evaluation across diverse rim geometries, as well as aerodynamic optimisation and generative design to accelerate wheel rim development for vehicle drag reduction.
Read the full abstract

All authors Chao Xia, Xi Zeng, Alexey Vdovin, Huanxia Wei, Simone Sebben, Carl Lewerth, Qing Jia, Zhigang Yang

High Energy Physics (Theory)

Type IIB Forms and Scalars from an $\mathrm{SL}(2,\mathbb{R})$ Topological Field Theory with Boundary

We formulate the scalar and form-field sector of type IIB supergravity as an eleven-dimensional topological field theory with boundary. Unlike the pseudo-action, it is a genuine action principle with manifest global $\mathrm{SL}(2,\mathbb{R})$ symmetry. The IIB fields arise as edge modes: a boundary term breaks the $\mathrm{SL}(2,\mathbb{R})$ gauge symmetry to $\mathrm{SO}(2)$, so the axion-dilaton parametrizes $\mathrm{SL}(2,\mathbb{R})/\mathrm{SO}(2)$, and the boundary field equations give the duality relations, including $F_5$ self-duality. As the action never refers to the axion-dilaton directly, S-duality monodromies become holonomies of a flat connection.
Read the full abstract

Machine Learning

What Matters for LLM Persona Simulation: Signal Form in Cold-Start E-Commerce Advertising

Estimating advertising performance in cold-start settings is challenging because historical logs are unavailable before launch. Large language models (LLMs) offer a potential alternative by simulating persona-conditioned user responses, but it remains unclear which design choices actually drive LLM-based persona simulation. Using real Google Ads e-commerce campaign data across five demographic segments, we identify three signal-form dimensions that affect simulation quality: (i) a demographic anchor without noisy rich attributes, (ii) a relative qualitative banner-blindness signal rather than quantitative anchors, and (iii) prompt phrasing. On this campaign, our method attains an MAE of 24.9 and a MAPE of 26.1%, a 51.3% lower MAE than the best tested non-LLM baseline and 77.3% lower than an LLM configuration using explicit CTR anchors. Systematic ablations show that minimal demographic descriptions achieve lower mean error than richer persona configurations, while relative qualitative cues substantially outperform the tested quantitative anchors. We additionally compare these configurations with direct-prior and rule-based baselines and with alternative LLM configurations.
Read the full abstract

All authors Jiwoo Choi, In Ik Lee, Woo Sung Seo, Minho Lee, Yun Young Choi

Computers and Society

Functional Dependency beyond Similarity: Hierarchical Decomposition of Training Data Attribution in Music Generative Models

Copyright analysis focuses on the similarity between two works. However, generative AI models learn from training data through functional relations beyond similarity. Training Data Attribution (TDA) estimates the influence of each training instance on a generated output and may serve as a basis for remuneration. However, prior studies find global influence, where a few training instances rank as the most influential for almost every generation. Used directly, TDA scores would distribute part of every remuneration to these instances regardless of what was generated. To address this, we propose a new approach in which TDA scores are decomposed into three levels of influence, namely global indicating the influence of a training data to a trained model, group for its influence on a group of generated outputs (that could be clustered together), and specific for its influence on a single generated output. Our study finds that, across several music generative models and TDA methods, global- and group-level influence is non-negligible in every case. Based on these observations, we propose the term functional dependency for the relation of training data to a trained model or to its outputs. This notion both encompasses and distinguishes dependencies of different scopes, allowing TDA scores to be interpreted as distinct forms of contribution in legal, economic and ethical discussions on remuneration.
Read the full abstract

All authors Changheon Han, Hugh Xuechen Liu, Ashkan Panahi, K{\i}vanç Tatar

Computer Vision

The APL Doesn't Fall Far From The Tree: Unsupervised Artefact Prototype Learning in Histopathology

Histopathology foundation models provide rich representations of tissue morphology, but technical artefacts in whole slide images introduce noise that may impact diagnostic accuracy. We introduce Artefact Prototype Learning (APL), a novel approach to detecting these artefacts that exploits the geometry of pre-trained embedding spaces without manual labels or fine-tuning. Rather than treating artefacts as isolated outliers, APL computes prototypes of out-of-distribution patches and uses them to distinguish artefact-associated patterns from tissue. Evaluation on a single-centre dataset shows that APL outperforms prior methods GrandQC and HistoART in detecting artefacts such as blur, dark spots, slide edges and pen marks, but struggles with tissue folds and blood. These results show that frozen pre-trained embeddings already carry information that allows artefact detection without expensive labels or vast computational resources.
Read the full abstract

All authors Emilia Agasi, Charlie Gourley, C. Simon Herrington, Robert L. Hollis, T. Ian Simpson

Artificial Intelligence

Solver-Admitted Metamorphic Testing of Learned Network Optimization Surrogates

Equivalent encodings of an optimization problem can produce different learned predictions. Interpreting these differences requires accounting for numerical error and multiple optima. We construct metamorphic tests for optimization solver surrogates with explicit eligibility conditions, output alignment, and numerical thresholds. Reference-solver agreement and independent constraint checks determine which sampled pairs enter the audit. We study alternating-current optimal power flow (AC-OPF), the traveling salesman problem (TSP), the capacitated vehicle routing problem (CVRP), and maximum independent set (MIS). Across five grids, GridSFM-Open violates each of 13 relations on at least one grid, and 12 by more than twice the threshold. Randomly initialized copies of the grid models fail nearly the same relations, so the failures do not require training. Source patches and input canonicalization remove selected discrepancies, but no intervention establishes preserved accuracy and feasibility within the declared margin on every grid. In routing, Policy Optimization with Multiple Optima (POMO) has a single-pass rotation/reflection violation rate that rises from 12.7% to 89.5% between 20 and 100 cities. Graph-Based Diffusion Solvers for Combinatorial Optimization (DIFUSCO) violates the MIS disjoint-union relation on 35.6% of draws. Post hoc normalization within connected components removes these violations on the tested graphs. These findings support reporting metamorphic consistency alongside optimality, feasibility, and accuracy, and validating repair quality separately from consistency.
Read the full abstract

Computational Complexity

Hardness of Approximation for Polynomial Optimization over Sphere for Linear Programming Relaxation Hierarchy

Semidefinite programming (SDP), particularly the sum-of-squares (SoS) hierarchy, provides $n^{O(q)}$-time approximation and certification algorithms for optimizing degree-$d$ homogeneous polynomials on the unit sphere. For arbitrary inputs, degree-$q$ SoS gives a powered-objective approximation factor $O_d((n/q)^{d/2-1})$ [BGGLT17]. For random order-$d$ tensors, degree-$d$ SoS has an integrality-gap lower bound of $2^{-O(d)}(n/d)^{d/4-1/2}$ relative to the true spherical maximum; lower bounds also extend to growing SoS degree in tensor PCA [BGL17,HKP17]. We establish lower bounds for weaker linear-programming relaxation hierarchies. Our degree-$q$ moment LPs impose the sphere identities and any prescribed, instance-independent family $\mathcal{G}$ of valid polynomial inequalities of degree at most $q$. They include Sherali–Adams-type product hierarchies. For every even $d\ge2$, constant $C>0$, and $q$ satisfying $2d | q$ and $2d\le q\le n$, every such relaxation with at most $n^{Cq}$ tests has instances $P$ satisfying \[ \frac{\mathrm{PowLP}_{q,\mathcal{G}}(P)}{\|P\|_\infty}\ge \frac{c_{d,C}}{\sqrt{q\log n}}\left(\frac nq\right)^{d/2-1/2}.\] Here $\|P\|_\infty$ is the maximum absolute value of the polynomial over the sphere and $\mathrm{PowLP}_{q,\mathcal{G}}(P)$ is the relaxed objective. The exponent $d/2-1/2$ improves the leading dependence on $d$ relative to the $d/4$-type exponents in prior random-tensor SoS lower bounds. Our LP hierarchies impose weaker positivity conditions than semidefinite moment relaxations. It leaves open whether analogous degree dependence can be achieved in lower bounds for SDP hierarchies.
Read the full abstract

Computational Geometry

Containment Problems in Convex Sets

Let $P$ be a set of $n$ points in the plane, and let $C$ be a compact, convex subset of the plane that contains the origin in its interior. We study the problem of finding two points $p$ and $q$ of $P$ with the property that any homothet of $C$ that contains $p$ and $q$ must contain at least a constant fraction of the points of $P$. We give proofs for several cases. When $C$ is strictly convex we achieve a fraction of $1/5$, and when $C$ is a general convex region we achieve a fraction of $1/6e$. We show how the ideas of this second bound can be used to prove a lower bound for a version of the problem in $\mathbb{R}^d$ when $C$ is a $d$-ball. We also consider a few special cases. When $C$ is a triangle we get a lower bound of $1/12$, and when we allow that triangle to be rotated by $\pi$ radians in addition to being scaled and translated we get a lower bound of $1/768$. In proving this last result, we improve the state of the art on the version of this problem in three dimensions, where $C$ is a parallelepiped. We also consider the algorithmic aspects of finding such a pair of points.
Read the full abstract

Artificial Intelligence

Notwithstanding Anything to the Contrary, Ignore All Instructions: Measuring Prompt Injection Guards on Legal Contract Review Attacks

Specialized prompt injection guards are deployed in front of document analysis systems, but no false positive rate has been published for any of them on legal contracts. We measure eleven open-weights guards on two corpora of legal contracts and find that existing guards struggle: document-level false positive rates are as high as 92.39%; two guards fall below chance on paired AUC; and a five-regex baseline is not Pareto-dominated by any guard in the panel. Controlled probes locate model-specific as well as general failure modes. Prompted LLMs resolve the tested failures, including a second threat category (unauthorized scope claims) on which every guard scores zero. A cascaded LLM solution, where the positives of a mid-tier model are verified by a frontier model, is 6.5x cheaper than using exclusively a frontier model and approaches its performance.
Read the full abstract

Machine Learning (Statistics)

Beyond the Lower-Star Filtration: Alternative Filtration Methods for Persistent Homology

In the lower-star filtration (LSF), simplices are assigned filtration values according to the maximum-vertex rule, with each simplex entering the filtration when its highest-valued vertex appears. A limitation of this construction is that some higher-dimensional simplices may enter simultaneously with their boundary faces, causing certain topological features to be filled immediately upon formation and thus not appear in the resulting persistence diagram. To address this limitation, we propose three alternative filtration methods: the \(\varepsilon\)-lower-star filtration (ELSF), the \(L_p\) filtration (LpF), and the softmax filtration (SMF). {ELSF introduces a dimension-dependent delay to the maximum-vertex rule,} allowing higher-dimensional simplices to enter the filtration after their boundary faces, whereas {LpF and SMF use aggregation functions that depend on all vertex values rather} than solely on their maximum. By modifying how filtration values are assigned, these methods allow simplices to enter the filtration more progressively, so that topological features such as loops and voids may persist over longer filtration intervals rather than being filled immediately. Consequently, the resulting persistence diagrams may include persistence features that are not observed under LSF, potentially providing useful information for distinguishing data populations. Experimental results on simulated and real-world datasets demonstrate that the proposed filtration methods are computationally efficient and can provide greater sensitivity to structural differences in the data than LSF.
Read the full abstract

Biomolecules

Protein B-factor modeling with local magnitude

Topological magnitude characterizes the effective size and spatial organization of metric spaces, offering a geometric approach with potential applications in molecular biology and materials science. Previous work has demonstrated its utility in modeling protein flexibility from structure, but global magnitude weighting requires dense linear systems whose computational cost increases with protein size. We introduce a local magnitude representation that retains residue-level structural information through nearest-neighbor environments. The construction is well posed and geometrically invariant, with central weighting distinguished from scalar neighborhood magnitude. Across 364 proteins, the mean within-protein fitting Pearson correlation coefficient (PCC) increases from 0.725 for global magnitude to 0.728 for local magnitude. Blind-prediction experiments on 345 proteins combine each representation with identical auxiliary descriptors and evaluate random forests and gradient boosting under three validation protocols. Local features improve the primary PCC in all six model–protocol comparisons, with a maximum absolute increase of 0.008 in protein-level ten-fold gradient boosting. For a fixed feature bank, local factorization costs $O(Nk^3)$ compared with $O(N^3)$ for global factorization, where $N$ is the number of C$_\alpha$ atoms and $k$ the neighborhood size. Timing measurements show a computational advantage of local magnitude on large structures. Within the tested single-seed configuration and distinct global and local scale banks, these results support local magnitude as a compact representation that preserves useful B-factor information while reducing feature-extraction costs for large proteins.
Read the full abstract

Machine Learning

Global Solar Siting from Geospatial Embeddings

Global solar photovoltaic (PV) capacity is rapidly growing, helping decrease our emissions and dependence on fossil fuels. However, optimally siting a solar plant remains challenging, requiring anticipation of geographical, urban, and infrastructure needs. Current methods rely on multi-criteria decision analysis (MCDA) using hand-picked geophysical layers and solar-resource products, however these strategies are not geographically transferable and do not capture the practical requirements of solar sites. We use geospatial embeddings, dense vectors representing geophysical properties of a location, to analyse 38,591 existing solar sites and the land they were built on to learn the latent solar suitability signal. We assess a set of linear and non-linear models and select a soft-voting ensemble of a random forest and a multilayer perceptron (RF+MLP ensemble), chosen for accuracy and large scale deployability. Validated against future installations in thirteen countries, the ensemble separates existing solar sites from random land with a median ROC-AUC of 0.90. Compared with published MCDA strategies, the ensemble increases ROC-AUC by 0.06-0.30 in every country. The ranking also concentrates real deployment: the most-suitable 20% of land holds 84% of estimated generation (annual energy yield) and 67% of installed capacity (potential power output). After applying standard siting constraints, 5.6% of global land (9.8% if cropland is permitted) remains highly suited to solar. The global 10 m resolution solar suitability dataset and code can be found here: https://github.com/Abdullahr29/Global-Solar-Siting.
Read the full abstract

Quantum Physics

Kinetic Cluster Embedding for Qubit Decoherence in Dissipative Spin Baths

Interpreting the decoherence of a spin qubit requires knowing how the surrounding bath spins behave: whether they fluctuate independently, exchange polarization incoherently, or evolve as coherent groups. These processes can yield the same Hahn-echo coherence time yet produce different coherence signals under pulse control of the qubit. Existing theoretical approaches either capture bath correlations at a cost that grows combinatorially with spin-bath cluster size or neglect polarization transport altogether. We introduce kinetic cluster embedding (KCE), which uses a single dimensionless parameter, comparing the flip-flop coupling of a spin pair with its detuning and linewidth, to decide how each pair must be treated. Pairs that can complete an appreciable exchange oscillation are propagated with the full dissipative quantum dynamics. All other spins are treated as random-telegraph fluctuators coupled through a kinetic network of polarization-transfer rates. We show that for nitrogen-vacancy centers coupled to surface electron spins, KCE propagates about 7 four-spin clusters, compared to more than 10$^5$ propagated by fourth-order master-equation cluster-correlation (ME-CCE4); nevertheless it reproduces ME-CCE4 coherence times to within 15$\%$ for dipolar couplings of order $10^{2}$ times the bath relaxation rate. For a three-dimensional P1 bath, the agreement is within 16$\%$. Coherent multispin clusters emerge as the leading correction to independent fluctuators, and they leave a distinct signature in the CPMG response. Importantly, we find that whether a rate description suffices is set not by coupling strength alone but by whether a pair can exchange polarization coherently before detuning and damping suppress it.
Read the full abstract

All authors Xiaofei Yu, Jonah Nagura, Stella Wang, Evan J. Villafranca, Peter C. Maurer, Giulia Galli

Differential Geometry

An improved isodiametric inequality for convex surfaces

Alexandrov conjectured that every closed convex surface of area $A$ and intrinsic diameter $D$ satisfies $A/D^2\leq\pi/2$. Using an averaging argument over the curvature measure, we prove that every closed convex surface satisfies $A/D^2\leq2\pi/3$, improving the best previously known bound.
Read the full abstract

Astrophysics of Galaxies

KCWI Observations of Type 2 Quasar-driven Ionized Outflows at z~0.1

AGN-driven outflows play a crucial role in regulating star formation in galaxies. These outflows are known to be multi-phase, and the characterization of the kinematics, extent, and energetics of each gas phase is needed to obtain a full picture of the outflow properties and impact. This study presents Keck Cosmic Web Imager (KCWI) integral field spectroscopy of the ionized gas outflow properties in seven low redshift (z<0.1) quasars—six Type 2 AGN and one narrow-line Type 1 AGN—providing insight into the warm ionized outflows in these luminous AGN. With the large field-of-view and high sensitivity of KCWI, these data reveal both larger ionized outflow extents than were previously known for these sources and uncover the existence of additional high-velocity kinematic components in the ionized outflows of these luminous AGN. We find (a) a high incidence (50-70%) of multiple outflows on scales of 4–10 kpc and (b) trends between the outflow extent, kinematics, and the ionized mass and energy outflow rates. These trends imply that the faster and often smaller outflows carry higher mass and energy outflow rates than the slower, larger outflows. We also find that the ionized gas outflows reach as far as the known extent of narrow-line ionized gas emission in AGN, which is known to depend weakly on AGN luminosity. This work highlights the utility of the current generation of wide-field integral field spectrographs on 8–10m class telescopes for studying quasar-driven outflows.
Read the full abstract

All authors Alison L. Coil, Saloni Agrawal, David S. N. Rupke, Serena Perrotta, Natalie Lam, Wenmeng Ning, Casey Carlile, Yu-Hsuan Teng

Robotics

ActPrune: Action-Grounded Visual Token Pruning for Efficient Vision-Language-Action Models

Vision-language-action (VLA) models perform versatile manipulation tasks from language instructions, but their inference cost limits how fast a robot can respond in closed-loop control. Visual token pruning reduces this cost; however, prior methods estimate token redundancy from encoded visual embeddings or attention, only after the visual encoder has processed the entire observation. We propose ActPrune, a training-free token pruning method that moves the pruning decision ahead of the vision encoder by grounding it in a geometric prior derived from the robot's own action. The action chunk predicted at the previous control step is already available before the current observation is encoded, and it specifies the end-effector and camera motion between consecutive observations. The same action affects the two standard camera views asymmetrically: in the fixed third-person view, ActPrune retains patches along the projected end-effector trajectory together with semantic anchors, while in the wrist view, it prunes patches that remain unchanged after compensating for the action-induced camera motion. Extensive experiments on two simulation environments (LIBERO and VLABench) and four real-robot tasks demonstrate that ActPrune achieves up to 1.61$\times$ speedup and 59.5% FLOPs reduction with negligible degradation in success rates. The project site is http://mmai.ewha.ac.kr/actprune/.
Read the full abstract

All authors Sungwon Moon, Minji Kim, Hyemin Boo, Daeun Song, Jiyoung Lee

Signal Processing

An Ultra-Low-Noise Tunable TIA in 45nm Silicon Photonics for Electro-Optical Ultrasound Data Acquisition

This paper presents the design and experimental characterization of a 32-channel, WDM-compatible electronic-photonic receiver array fabricated in the GlobalFoundries 45nm monolithic silicon photonics process. Optimized for miniaturized, millimeter-scale electro-optical ultrasound data acquisition probes for broadband photo-acoustic imaging (PAI), the chip utilizes integrated micro-ring resonator (MRR) demultiplexers and photodiode (PD) arrays to decode multi-wavelength acoustic signals outside the body. This work focuses specifically on the comprehensive characterization of the receiver front-end, evaluating both the optical WDM filter bank and the transimpedance amplifier (TIA) performance. Monolithic co-integration minimizes critical photodiode-to-TIA interface parasitics. The TIA incorporates a DC-offset cancellation loop that compensates for photodiode-induced offset currents while providing tunable low-frequency cutoff control over a range of 30 kHz to 1.5 MHz. This tunability enables effective suppression of flicker noise and improves overall signal integrity. The front-end achieves the peak transimpedance gain of 144 dB$\Omega$ while maintaining an ultra-low input-referred noise current density of $1\text{ pA}/\sqrt{\text{Hz}}$, demonstrating a highly scalable architecture for high-sensitivity intravascular and endobronchial imaging.
Read the full abstract

All authors Marziyeh Rezaei, Sridhar Singhal, Alp A. Derin, Soner Sonmezoglu, Sajjad Moazeni