Peer Revenue Returns subject to peer review

Papers

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

All of arXiv

Wednesday 2 September

High Energy Physics (Theory)

Unfolded Krylov complexity: universal chaotic dynamics without false positives

A central challenge in diagnosing quantum chaos is to distinguish genuine many-body scrambling from kinematic effects of the spectrum. Krylov state complexity, or spread complexity, has emerged as a powerful diagnostic, with its characteristic growth, peak, and relaxation often taken as signatures of chaos. However, previous work has shown that this criterion can give false positives: saddle-dominated integrable systems may display prominent peaks even without random-matrix level correlations. We argue, based on complementary numerical and analytical evidence, that this ambiguity can be resolved by unfolding the spectrum prior to constructing the ensuing Krylov dynamics. By removing the non-universal smooth density of states while retaining microscopic spectral correlations, unfolding suppresses spurious peaks in integrable systems while preserving the universal spectral signatures of chaotic systems. Analytically, the formulation of the Lanczos iteration in terms of orthogonal polynomials yields an exact complexity kernel with a robust near-diagonal structure whose fine-grained features reflect the underlying spectral correlations. Moreover, for the logarithmic model, unfolding can be performed exactly, mapping the spectrum to a uniform lattice and yielding an analytic spread complexity that removes the false-positive peak. These findings establish unfolded Krylov complexity as a more reliable probe of genuine many-body scrambling.
Read the full abstract

All authors Johanna Erdmenger, Kyoung-Bum Huh, Hyun-Sik Jeong, Juan F. Pedraza

Distributed Computing

Towards Global Federated Genome-Wide Association Meta-Analysis Using GA4GH TES

Genome-wide association studies (GWAS) gain statistical power from large, ancestrally diverse cohorts, but privacy regulations and data-residency constraints often prevent genomic data from being centrally pooled across institutional or national borders. We present a privacy-preserving federated GWAS meta-analysis pipeline built on the APPFL framework, in which each site computes local GWAS summary statistics and transmits only aggregate results, never individual-level genotypes. Analysis is executed through a global network of Global Alliance for Genomics and Health (GA4GH) Task Execution Service (TES) endpoints, which allows computation to move to the data rather than the reverse. The server performs inverse-variance-weighted fixed-effect meta-analysis and returns aggregated results to all sites, while HiveWatch, our developed geographic observability toolkit, provides real-time monitoring of distributed task execution. In a five-site simulation over 100,000 synthetic individuals and roughly 240,000 variants for Type 2 Diabetes and Body Mass Index, the federated meta-analysis reproduces the association signal expected from a pooled analysis without centralizing any genotype data, showing that standards-based task execution and federated learning enables a practical privacy-preserving infrastructure for international GWAS meta-analysis.
Read the full abstract

All authors Abhijit Chunduru, Matthew Joel, Zilinghan Li, Ravi Madduri

Quantum Physics

Coherent microwave-to-optical transduction with Yb:YSO spins strongly coupled to a 3D resonator

Microwave-to-optical quantum transducers are essential for entangling remote superconducting qubits. Among the available transduction platforms, ensembles of Er$^{3+}$ and Yb$^{3+}$ ions doped into solids have emerged as leading candidates. While external magnetic fields are needed to split the Zeeman levels of erbium ions and enable a microwave–qubit interface, superconducting qubits suffer decoherence in such fields. In contrast, ytterbium ions exhibit zero-first-order Zeeman transitions and large hyperfine splittings at zero magnetic field (when doped into inorganic crystals). Owing to its long optical and spin coherence times, Yb:YSO has been widely used as a quantum memory, yet its potential for quantum transduction remains largely unexplored. Investigating this material could enable the integration of quantum memory and transduction in a single platform. Here, we demonstrate microwave-to-optical transduction in the continuous-wave regime using a 5\,ppm doped Yb:YSO crystal. The internal transduction efficiency is $2\times10^{-8}$ with a bandwidth of 200\,kHz, achieved using a 3D loop-gap microwave resonator (LGR) and a single-pass optical configuration. We explore all the ground states that form a V-type three-level system with the first and second optical excited states and assert the use of the ground state, which provides the highest efficiency and isolated optical transition. We further establish strong spin-microwave coupling from avoided crossing measurements. With a strong microwave drive to saturate the spin transition, we estimate the spin population pumped into the excited state, close to the simulated value. Finally, we calculate target parameter values for maximum efficiency with our system and suggest using 50\,ppm doped Yb:YSO crystal. With the calculated target parameters, the internal transduction efficiency is predicted to reach up to $10^{-4}$ in the current LGR.
Read the full abstract

All authors Ujjwal Gautam, Nasser Gohari Kamel, Sourabh Kumar, Daniel Oblak

Analysis of PDEs

The direct moving plane method for weak solutions of the fractional $p$-Laplacian

In this paper, we develop the method of moving planes entirely in the weak formulation for the fractional $p$-Laplacian in the singular range $1<p\leq2$. We first establish a small region principle for antisymmetric functions and apply it to prove radial symmetry and monotonicity of nonnegative weak solutions of fractional $p$-Laplacian equations in a bounded domain. We also consider the nonlocal quasilinear Lane–Emden equation $(-\Delta)^s_pu=u^q$ in $\mathbb R^n$. In the Sobolev critical case, we establish radial symmetry, monotonicity, and precise asymptotic behavior at infinity for finite-energy weak solutions. Under a suitable decay condition, we also obtain radial symmetry for the full range $q>p-1$. Our proofs rely on integral estimates derived directly from the weak formulation, thereby extending the moving plane approach of Chen and Li (Adv. Math. 2018: 735-758) from the pointwise $C^{1,1}_{\mathrm{loc}}$ setting to weak solutions.
Read the full abstract

Computer Vision

Signal or Noise? Auditing Rotation-Induced Saliency Drift in Medical and Aerial Imaging

Post-hoc saliency maps such as Grad-CAM are increasingly used to audit why a deployed vision model made a decision, yet the heatmap drifts when the input is rotated, even when the prediction is unchanged. In domains with no canonical orientation, such as histopathology and aerial imagery, this undermines using saliency as evidence. We ask whether that drift is faithful signal or noise introduced by the CAM operator, and answer it by measuring equivariance at every stage of the operator rather than inferring it from the network's output. The instability is not where one would guess: the channel weights are the most rotation-stable stage, and on ResNet-50 exactly stable, because a GAP+linear head makes the class gradient field spatially constant. What moves is the spatial activation tensor, and the classifier's own pooling discards that movement. A causal test confirms the consequence: occluding the pixels whose saliency drifts costs the model less than occluding random pixels, at either orientation. The drift is carried by degrees of freedom the classifier throws away, which is what makes removing it faithful rather than destructive. EquiGrad-CAM is a training-free wrapper that takes T rotated views, inverse-rotates each view's saliency into a common canonical frame, and averages. On the full ImageNet-1K validation set it raises equivariance over single-view Grad-CAM by +36.0% (ResNet-50), +87.5% (VGG-16) and +247% (ViT-B/16); a scale-matched ablation isolates alignment before averaging, not the locus of aggregation, as the driver. It beats rotation-augmented training without retraining, lifts zero-shot CLIP by +145%, and yields rotation-consistent explanations on PatchCamelyon and RESISC45. Its by-product PEUM ranks explanations by how reproducible they are, at no cost beyond the views already taken. Code: https://github.com/Khawaja-Murad/EquiGrad-CAM
Read the full abstract

General Relativity and Quantum Cosmology

Accretion dynamics of thin and thick disks in charged Kalb-Ramond black holes

In this work, we investigate the accretion properties of an electrically charged, static, spherically symmetric black hole in Kalb-Ramond (KR) gravity, where a dimensionless parameter $l$ characterizes spontaneous Lorentz symmetry-breaking. We first constrain the free charge and KR parameter space, $(Q,l)$, by comparing the predicted shadow radius of this family of solutions with the bounds inferred from Event Horizon Telescope observations of Sagittarius A*. The inclusion of electric charge introduces a degeneracy between $Q$ and $l$, enlarging the region compatible with the observed shadow size and yielding the broad constraint $-0.3900 \leq l \leq 0.1898$, with part of the compatible parameter space extending into the naked-singularity sector. We then investigate relativistic accretion disks around the charged KR black hole within the above constraints. For geometrically thin disks, we compute the time averaged energy flux, temperature distribution, emission spectrum, time averaged torque, and mass-to-radiation conversion efficiency, finding that the radiative output increases for higher values of $l$, while lower values of $l$ suppress it. Furthermore, when coupled with the electric charge this effect is amplified. For geometrically thick, constant-angular momentum tori-shaped disks, Lorentz symmetry-breaking shifts the characteristic orbital radii, compresses the radial extent of equilibrium configurations, narrows the interval supporting finite equilibrium tori, and shifts the stationary-fluid topology toward open and unconfined configurations outside that interval. These results show how charge and spontaneous Lorentz symmetry-breaking jointly affect observational constraints and the structure and emission properties of black hole accretion flows, providing potential signatures of KR gravity in the strong-gravity regime.
Read the full abstract

All authors Ednaldo L. B. Junior, José Tarciso S. S. Junior, Francisco S. N. Lobo, Jorde A. A. Ramos, Manuel E. Rodrigues, Diego Rubiera-Garcia, Luís F. Dias da Silva, Henrique A. Vieira

Robotics

FOCUS: Foot Observation Confidence for Robust Humanoid Proprioceptive Odometry

Foot forward kinematics (FK) is widely used to improve proprioceptive legged odometry by providing reliable velocity constraints during foot support. Existing contact-aided estimators generally rely on binary contact decisions to determine whether the FK measurements of an entire foot should be trusted. However, contact does not necessarily imply FK reliability. Dynamic locomotion often involves partial support, toe dragging, and foot slip, causing binary contact decisions to accumulate significant drift over long trajectories. To address this limitation, we propose FOCUS (Foot Observation Confidence from Unannotated Simulation), which predicts a continuous FK reliability weight for each foot instead of estimating binary foot contact. Rather than replacing the model-based estimator, the predicted reliability weights are used to blend FK velocity observations with IMU-propagated body velocity and to adapt the observation covariance of an extended Kalman filter (EKF), enabling smooth reliability-aware fusion without hard contact switching. The network is trained from automatically generated simulation signals using an FK-weighted velocity consistency loss with lightweight simulator-contact regularization, without manually annotated continuous FK-reliability labels. The deployed model relies only on IMU and joint kinematic measurements, making it suitable for hardware platforms with unreliable torque sensing. Experiments demonstrate that FOCUS reduces absolute trajectory error (ATE) by 83.7% on simulated walking episodes, preserves simulated dynamic-motion fidelity in motion scale and spectral energy, reduces ATE by 70.8% across 19 real walking segments, and reduces mean ATE by 42.7% across four real dynamic-motion routines.
Read the full abstract

All authors Kaixin Feng, Angsong Li, Shaopeng Zhang, Enyu Li, Peiwen Lin, Chuang Wang, You Li, Haiyu Lan

Instrumentation and Methods for Astrophysics

Demonstrating the Time-Domain Capabilities of the 4-m International Liquid Mirror Telescope: An Early Census of Transient and Variable Detections

The PyLMT transient detection pipeline has been operational since November 2023, detecting transient and variable objects in the ILMT images in almost real time. Using the image subtraction technique, nearly 3700 CCD frames have been analyzed by the automated pipeline, generating nearly 23,000 alerts for the detection of verified transient candidates and cataloged variable sources. Around 21,000 of the alerts correspond to known MPC asteroids, nearly 2000 correspond to variable stars (including eclipsing binaries, RR Lyrae, Delta Scuti, T-Tauri, etc.), 509 correspond to variable AGNs (including QSOs, Seyfert galaxies, and blazars), 21 supernova candidates, and several other interesting candidates. We provide a concise overview of the detections and their significance, emphasizing the surveys potential contributions to a broad class of astrophysical and scientific cases. A transient detection dashboard called DART was developed using Streamlit to visualize and categorize candidates based on PyLMT and SIMBAD classifications. It includes cone-search functionality and displays key metadata, offering an intuitive interface that is publicly accessible. Our results demonstrate the viability of liquid mirror telescopes such as the ILMT for time-domain astronomy, emphasizing the important role that small-field survey facilities can play in systematic transient science programs.
Read the full abstract

All authors Kumar Pranshu, Kuntal Misra, Bhavya Ailawadhi, Monalisa Dubey, Naveen Dukiya, Sara Filali, Paul Hickson, Priyanshi Kumari, Gokul Singh Mehra, Vibhore Negi, Jeewan C. Pandey, Anna Pospieszalska-Surdej, Jean Surdej, Sarvesh Kumar Yadav

Combinatorics

Recurrences for permutations with long increasing subsequences

We prove two simple bivariate recurrences for the number of permutations with a long increasing subsequence. The two recurrences imply D-finiteness of the sequence in a certain range. As a consequence, we also obtain a proof of a conjecture posed by Kauers and Koutschan in 2023.

Robotics

Hardware-Accelerated Instance Segmentation for Resource-Constrained Space Robotics with Criticality Analysis

Autonomous lunar missions require real-time per- ception under three coupled constraints: extreme low-light conditions, limited onboard compute, and radiation-induced hardware faults that can silently corrupt inference. We present a deployment-oriented instance segmentation framework for resource-constrained lunar robotics that jointly addresses quan- tization calibration and system-level fault exposure under strict compute constraints. First, we introduce Activation Variance Informative Sampling (AVIS), a label-free calibration strategy that deterministically selects calibration samples based on activation variance statistics. Second, we deploy a YOLO-based segmentation model on a Deep Learning Processor Unit (DPU) with architectural modifications that reduce CPU fallback paths and enable statically compiled execution with bounded latency in low-lighting conditions. We further introduce a software-level criticality analysis to estimate fault exposure and guide mitigation under radiation-constrained operation. On a lunar micro-rover platform, AVIS with bias correction recovers 69.8% of quantization-induced accuracy loss while achieving 309 ms inference latency and 5.7 W power consumption. Targeted mitigation reduces global criticality by 31.7%. The results demonstrate an integrated approach and a blueprint for a reliable and safe AI perception framework under space deployment constraints.
Read the full abstract

All authors Siddhant Shete, Hilmi Dogu Kücüker, Udo Frese, Frank Kirchner

Combinatorics

Neumaier graphs of coherent rank five

We construct an infinite family of Neumaier graphs of coherent rank five, answering the existence question at the smallest possible coherent rank beyond the strongly regular case. For every prime power $q\geqslant7$ with $q\equiv3\pmod4$, set $n=q+1$. Each graph in our construction has precisely five distinct eigenvalues, Neumaier parameters \[ \left( n(n-1)(n-3), \frac{n^2(n-3)}2, \frac{n(n^2-n-8)}4; \frac{(n-2)^2}{2}, (n-1)(n-3) \right), \] and its adjacency matrix lies in the Bose–Mesner algebra of the four-class association scheme of Holzmann, Kharaghani, and Suda. Paley Hadamard matrices and Desarguesian mutually orthogonal Latin squares yield an infinite family whose smallest member has parameters $(280,160,96;18,35)$.
Read the full abstract

Artificial Intelligence

SkillGLoW: Procedural-Family Skill Consolidation for Self-Improving Agents on Long-Horizon Task Streams

LLM agents increasingly self-improve by writing and reusing textual skills, kept either as one global document or as a flat pool of per-task entries, though most of the evidence comes from domains with structurally similar tasks. On long-horizon workloads where each task demands a different solution, the two forms fail in opposite ways: the document collapses into generic discipline, while the pool inflates and its entries stay bound to the instance that wrote them. We argue the missing unit of reuse is the solving procedure shared by a cluster of related tasks, and build SkillGLoW (Global-Local Weave) around it: the local skills a task writes from its own execution are aggregated into procedural families and compressed into de-instantiated global priors, while the instance detail they hold is regenerated per task rather than stored; a commit gate admits a prior only when real execution shows it does not degrade the deployed library. Across four benchmarks (mathematical reasoning, terminal automation, software repair, and embodied control) and three models, the priors gain 17.2 points (hard) over the no-skill baseline on average, with positive gains in all 12 continual-improvement runs, and 18.0 with local regeneration, while the library holds one prior per procedural family, 3.6x more compact than the per-task pool. Under the same protocol GLoW leads a published single-document optimizer on 15 of 21 cells. Unmodified, the library lifts success on unseen ALFWorld tasks from 73.9% to 83.9%, evidence that what transfers is procedure rather than task memory.
Read the full abstract

All authors Ao Yan, Xin Zhang, Jiawei Du, Joey Tianyi Zhou

Artificial Intelligence

PEARL: Path-Entity Aligned Relational Learning with Contextual Subgraphs for Inductive Knowledge Graph Completion

Inductive knowledge graph completion (IKGC) aims to predict missing links involving entities unseen during training, requiring models to learn transferable relational and structural patterns. Existing subgraph- and path-based approaches often encode relational paths independently of their surrounding query subgraphs, although the predictive relevance of a path may vary across structural contexts. We propose PEARL, a Path-Entity Aligned Relational Learning framework that models paths as context-conditioned reasoning signals. PEARL constructs a query-specific contextual subgraph from the union of the query entities' neighborhoods and uses a large language model (LLM)-guided retriever to distill semantically relevant paths. It then builds a bipartite interaction graph over paths, contextual entities, and a global subgraph representation, allowing path embeddings to adapt to local and global structural evidence. To suppress noise introduced by the enlarged context, PEARL employs a dual-view contrastive objective that promotes representation consistency under stochastic contextual perturbations. Experiments on WN18RR, FB15k-237, and NELL-995 show that PEARL obtains the best average Hits@10 among the compared IKGC methods on all three benchmarks. Ablation studies, efficiency analyses, and case studies further validate the contributions of contextual subgraph modeling, semantic path retrieval, path-entity interaction, and contrastive regularization.
Read the full abstract

Artificial Intelligence

ASCII Attack: Recontextualising Harmful Requests as Artistic Critique in Large Language Models

Safety alignment trains large language models to refuse harmful requests stated plainly, but that training is applied mostly to surface form. Requests that only recontextualise the same operational content, changing how the model reads it, are therefore only weakly covered. The ASCII Attack is one such recontextualisation. It is single-turn and black-box: one message, with no access to model internals. It embeds a fully legible harmful request in ASCIl-art characters, presents it as artwork, and asks for feedback. Unlike ArtPrompt, it hides nothing: the request stays readable. The reply is written as artistic critique and can contain operational detail that a plain request would have been refused for. Every framed prompt is paired with a direct-question control, so the contrast is isolated from topic, model and decoding variation. The contrast identifies a bundled surface, not one isolated channel. Across eleven models and eight harm topics, a harm-aware classifier judges 62% of framed prompts harmful against 42% of controls. On the most susceptible model the framed prompt succeeds 93% of the time. A single query matches or exceeds published single-query attacks under four of five harm judges. The effect tracks the model more than the topic and does not diminish with scale. At least one judge dissents from the panel majority on nearly two-thirds of framed rows, which is itself a measurement-validity finding. That pattern is consistent with mismatched generalisation.
Read the full abstract

All authors Da Cheng Gu, Yifei Dong, Xinghao Yang, Yongshun Gong, Wei Liu

Statistics Theory

Tests of graph homogeneity, subgraph counts, and quasirandomness

Homogeneous random graphs, also known as Erd{\H os}-Rényi graphs, are a subset of the family of dense random graphs, specified by a graphon. We analyze several goodness-of-fit tests for these models that are based on subgraph counts. To obtain the limiting null distribution of the test statistics we use a decomposition of graph functionals, which reveals a cancellation effect. Motivated by a quasirandomness result we obtain a test that is consistent against all alternatives, and we use two popular parametric subfamilies to evaluate the other tests. The theoretical results refer to the limit $n\to \infty$ of the size $n$ of the graph, the behavior for finite $n$ is illustrated by simulations.
Read the full abstract

Functional Analysis

Invariant subspaces for free linearizations of Lipschitz maps

We consider the invariant subspace problem for operators induced by Lipschitz self-maps on Lipschitz-free spaces. Besides the usual free linearizations $\widehat f$ of basepoint-preserving Lipschitz self-maps, we consider a wider class of operators $T_{f,e}$ which are naturally defined for arbitrary Lipschitz self-maps. We show, among other results, that every $\widehat f$ admits a non-trivial invariant subspace whenever the underlying metric space contains a compact ball, or has at least two connected components one of which has non-empty interior. We also obtain corresponding positive results for $T_{f,e}$ under isolated-point, compact-ball and disconnectedness assumptions, and discuss some consequences for linear dynamics.
Read the full abstract

All authors Clément Coine, Pedro L. Kaufmann, Colin Petitjean, Sebastián Tapia-García

Computer Vision

FuDU: A Fuzzy Dual-dimensional Uncertainty Framework for Streaming Active Learning in Industrial Defect Detection

Ensuring the reliability of deep learning models in real-time industrial defect detection is critical for high-stakes quality inspection. To mine uncertain samples within continuous industrial media streams, thereby enhancing the reliability of the detection system, this paper proposes a streaming active learning method based on the Fuzzy Dual-dimensional Uncertainty (FuDU) framework. Specifically, we first design a Prototype-based Global Uncertainty Quantification (PGUQ) module on the backbone to evaluate image-level uncertainty via normal/defective feature prototypes. A Dual-entropy defect Uncertainty Evaluator (DeUE) is then integrated into the detection head to quantify box-level uncertainty. Finally, by modeling uncertainty as systematic error, we propose a fuzzy dual-dimensional uncertainty-aware strategy that leverages fuzzy inference to fuse dual-dimensional uncertainties, enabling expert knowledge-driven adaptive sampling decisions. Comprehensive experiments demonstrate that FuDU is efficient and flexible, making it well-suited for challenging industrial inspection tasks such as the detection of nuclear fuel rod defects. Our code is publicly available at: https://github.com/wangzhaoyang-508/FuDU.
Read the full abstract

All authors Zhaoyang Wang, Haiyong Chen, Binyi Su, Xinwei Lyu

Solar and Stellar Astrophysics

The relationship between solar and stellar tachoclines, dynamos, and spin-down

The solar tachocline, a thin shear layer separating the differentially rotating convective zone (CZ) from the underlying rigidly rotating radiative zone (RZ), remains a dynamical mystery. Under the influence of "radiative spreading" (the process by which meridional circulation, baroclinicity, and differential rotation burrow through a stably stratified fluid), the tachocline should have spread significantly by the current age of the Sun. Some unknown torque must therefore rigidify the whole solar interior below the CZ and keep the tachocline confined to a thin layer. At the same time, the spin-down process (through which cool stars shed angular momentum via surface magnetic torques) must extract angular momentum from the solar RZ, presenting a second mystery: solar spin-down must be communicated from the near-surface layers to the deep interior, all the while leaving the tachocline intact. In this work, we explicitly analyze the dynamics of radiative spreading in two 3D, spherical-shell, fully nonlinear fluid simulations of a solar-like CZ–RZ system, one without a magnetic field (hydrodynamic—HD) and one with a small random seed magnetic field (magnetohydrodynamic—MHD). We find that radiative spreading is unmitigated in the HD case (as expected), but in the MHD case, a self-excited dynamo not only confines the tachocline, but also extracts angular momentum from the RZ, thereby communicating the spin-down downward. We thus speculate that solar and stellar tachoclines may be intimately linked to both the spin-down process and global dynamo.
Read the full abstract

Computer Vision

Asymmetric Paired-Annotation Learning for Multi-Structure ULF Pediatric Brain MRI Segmentation

Portable ultra-low-field (ULF) MRI can expand access to pediatric neuroimaging, but segmentation at 0.064 T remains challenging because anatomical boundaries are weakly delineated, small structures may be only partially visible, and high-field references can be locally misregistered. The LISA 2026 Challenge provides two non-equivalent annotations reflecting different sources of anatomical evidence: a highfield-derived (HF) mask defining the scored target and a low-field-edited (LF) mask aligned with visible ULF anatomy. In this challenge report, we describe AURA, an nnU-Net-based asymmetric supervision strategy that treats these annotations as distinct observations rather than interchangeable ground truths. AURA anchors training to the HF mask and incorporates the LF mask through a bounded reliability gate based on label disagreement, boundaries, predictive uncertainty, class reliability, and training stage. On a 16-case development split, the HF-supervised baseline, AURA, and their ensemble achieved Dice scores of 0.7984, 0.7950, and 0.7988, respectively, while the ensemble achieved an HD95 of 1.8892 and an ASSD of 0.7855. These results provide a preliminary evaluation of AURA within the LISA 2026 Challenge and motivate further assessment on the hidden test set and external ULF cohorts. Our code and pretrained models are available at https://github.com/minhdang050806/ A-nnU-Net-based-asymmetric-supervision-strategy.
Read the full abstract

All authors Ha-Hieu Pham, Dang P.M. Cao, Minh Hoang Pham, Khanh Nguyen Vo Ngoc, Thanh-Huy Nguyen, Ulas Bagci, Huy-Hieu Pham

Chemical Physics

Prototype-guided transfer of sparse literature knowledge for electrolyte additive discovery

Electrolyte additive discovery remains challenging because experimentally validated molecules are sparse, whereas accessible chemical spaces are vast and largely unlabeled. This challenge is amplified in lithium-ion batteries, where additive performance arises from coupled interfacial reactions rather than a single molecular property. Here, we develop a prototype-guided molecular intelligence, ProtoMI, a literature-driven framework that learns transferable structural priors from reported electrolyte additives and uses them to prioritize candidates in unlabeled chemical space. For boron-containing additives, ProtoMI combines 126 literature-reported molecules with 179,977 unlabeled candidates. Graph contrastive learning identifies seven chemically interpretable prototypes from the reported additives, and prototype guided semi-supervised contrastive learning adapts these prototypes to the candidate space under source-target distribution mismatch. In retrospective temporal validation, ProtoMI achieves enrichment factors of 9.2-45.6 while screening less than 2% of the candidate space. A subsequent translation step identifies four commercially accessible candidates. One representative candidate, 4,4,5,5-Tetramethyl-2-[10-(1naphthyl)anthracen-9-yl]-1,3,2-dioxaborolane (TNDB), improves high-temperature LiFePO4||graphite cycling at 55 {\deg}C by 34.93% relative to the baseline electrolyte. An arsenal of characterizations and operando optical fiber Fourier transform infrared spectroscopy suggest that TNDB forms B-containing, F/P/O-modified inorganic interphases, suppresses solvent decomposition and reduces Fe deposition on graphite. This case study shows how sparse literature knowledge can guide experimentally efficient molecular discovery in data-scarce battery-additive spaces.
Read the full abstract

All authors Weixiang Hong, Hongting Du, Jiayue Tang, Ruifeng Tan, Yangjian Quan, Jia Li, Jiaqiang Huang

Cryptography and Security

Agentic Settlement Protocol: An Application Profile for Refundable, Delayed-Fulfilment Agent Commerce on Stablecoin Rails

Autonomous agents can already pay per request: HTTP-native protocols such as x402 let an agent sign a stablecoin authorization and receive a resource in the same round trip. That model is atomic and final, which suits metered access and fails commerce: a purchase made on a person's behalf – a service appointment, a physical order, a flight – is large, frequently cancelled, and should not become the seller's money until delivery. We present the Agentic Settlement Protocol (ASP), an application profile over on-chain authorize-and-capture escrow (as standardised by the Commerce Payments Protocol) for businesses whose fulfilment is confirmed off-chain by their own order, scheduling, invoicing or booking system, abstracted as a fulfilment engine. ASP contributes: a three-deadline hold model separating the issuance deadline, escrow expiry and the engine's inventory expiry by an explicit submission-inclusion-finality margin, under which no inventory is issued against reclaimable funds; a fulfilment-verification ladder stating who is trusted to trigger capture, what their attestation proves, and under what challenge window; engine-authoritative partial refunds with a refund-liquidity order and per-seller exposure controls, including atomic exposure reservation, that bound the operator's credit risk; distributor revenue share that is provably unprofitable to self-deal; a single-currency-per-charge invariant; and a normative interface specification (x402 scheme, vault ABI, operator and connector APIs, conformance levels) intended to let independent implementations interoperate. The design originated in a review of travel-agency participation in agentic settlement and is instantiated on the XDC Network. This is a design paper; a fault-injection evaluation plan is specified and measurement is left to follow-up work.
Read the full abstract

Computer Vision

LeakageBench: Document-Level Leakage Risk for Redacting Personally Identifiable Information in Document Images

Real-world personally identifiable information (PII) redaction often operates on document images—scans, screenshots, and PDF renderings—where OCR errors, layout structure, and visual noise determine whether sensitive information is actually removed. Existing PII benchmarks are mostly text-centric and do not measure document-level redaction risk: a page remains unsafe if even one identifier is missed. We introduce LeakageBench, a challenge set of 500 document images with 11,954 GDPR-aligned PII annotations spanning direct identifiers, linkage keys, and contextual re-identification surfaces. We evaluate generic OCR pipelines, commercial and task-adapted OCR-dependent detectors, and OCR-free vision-language models using entity-level F1, group-wise leakage, and document-level leakage metrics. Code Interpreter raises GPT-5.5 localization F1 from 0.090 to 0.249, but critical page-level leakage remains 0.968. These results show that stronger detection and tool assistance improve localization without making most pages safe for release. LeakageBench provides a diagnostic benchmark for high-recall, spatially grounded PII redaction in document images.
Read the full abstract

Materials Science

Decoding Oxygen K-edge Fingerprints of NCM-811 Degradation via Ab Initio Many-Body Theory and High-Throughput Screening of Crystal Proxies

The degradation of LiNi$_{0.8}$Co$_{0.1}$Mn$_{0.1}$O$_2$ (NCM-811) in Li-ion batteries produces complex transition-metal oxides and binary phases that fundamentally limit cathode performance. While identifying these degradation products via X-ray absorption spectroscopy (XAS) is essential for mitigating electrochemical performance loss, interpretation remains challenging due to the structural complexity of real-world samples. In this work, we present an integrated theoretical-experimental framework combining high-resolution oxygen K-edge XAS with ab initio simulations based on many-body perturbation theory and high-throughput screening from density functional theory. We first evaluate the spectroscopic fingerprints of eight layered, spinel, and nominal rock-salt reference oxides, identifying discrepancies between the idealized single-crystal bulk phase and experimental spectra. Using high-throughput screening to analyze the oxygen $p$-projected density of states of 38 distinct polymorphs of NiO, CoO, and MnO, we propose that the spectral differences can emerge, among other factors, from local structural variations represented here by simplified structural proxies. Our work establishes a viable and rigorous computational pathway to interpret the complex landscape of degraded battery materials.
Read the full abstract

All authors Daniel Duarte-Ruiz, Timo Reents, Elmar Kataev, Raul Garcia-Diez, Regan G. Wilks, Marcus Baer, Caterina Cocchi

Signal Processing

A Geometric Analysis of Initialization Bias in Spherical $K$-means in the Weak Signal Regime

We study initialization bias in spherical $K$-means for weakly informative directional mixtures. We model the observations by a $K$-component von Mises-Fisher mixture with a small concentration parameter $\kappa$, corresponding to a high-dispersion regime in which the data provide limited information about the underlying directions. Our analysis begins with the limiting case $\kappa=0$ (corresponding to a uniform distribution over the sphere), where one population spherical $K$-means update is governed entirely by the Voronoi tessellation induced by the initialized templates. For uniformly random initializations in fixed dimension $d$, the updated templates become asymptotically aligned with their initial values as $K\to\infty$: the average squared geodesic error scales as $O(K^{-2/(d-1)})$, while the worst-case error is $O((\log K/K)^{2/(d-1)})$. We then show that, in the weak-signal regime of small positive $\kappa$, the population update remains an $O(\kappa)$ perturbation of this limiting map. Thus, in the weak-signal regime, spherical $K$-means can preserve initialization-induced structure despite the presence of a genuine but highly dispersed directional signal.
Read the full abstract

Computer Vision

TAME: Temporal-Aware Mixture-of-Experts for Text-Video Retrieval

Text-Video Retrieval (TVR) retrieves videos that match a natural-language query, but extending image-text models such as CLIP to videos is fundamentally limited by the lack of temporal modeling. Videos exhibit frame-wise heterogeneity in appearance and motion, and compressing all frames into a single representation often obscures temporal structure and semantic transitions. To address this, we propose Temporal-Aware Mixture-of-Experts for Text-Video Retrieval (TAME), a CLIP-based framework that jointly models frame-level structure and temporal relations. First, we integrate sparse Mixture-of-Experts (MoE) layers into both CLIP encoders and apply frame-consistent routing on the vision branch so that experts specialize according to frame-level visual patterns while preserving the original vision-language alignment. Second, we introduce Frame-Temporal (FT) tokens that aggregate global cross-frame information and feed it back to each frame, enabling the visual encoder to capture long-range temporal dependencies without harming local details. Third, we design a Cross-Temporal Interaction and Aggregation (CTIA) module that refines frame-wise sentence-video similarities through staged temporal filtering and fusion. Experiments on standard TVR benchmarks show that TAME consistently improves over CLIP-based baselines. On MSR-VTT, it improves R@1 by 4.0 over CLIP4Clip, and also achieves consistent gains on DiDeMo, MSVD, LSMDC, and ActivityNet. The code is available at https://github.com/sejong-rcv/TAME.
Read the full abstract

All authors Uicheol Jung, Juyoung Hong, Hojung Kwon, Yukyung Choi

Machine Learning

SMart: A Multi-source Multi-phase Time Series Representation Transfer Framework

Time series representation learning (TSRL) has attracted growing research interests in recent years. Two recent explorations in TSRL are: i) exploiting a transformer-based framework to learn time series; ii) instead of using only the targeted dataset, borrowing time series from other datasets to to facilitate representation transfer. While these two explorations are shown effective, the self-supervised time series recovery task in (i) and the single-source dataset used in (ii) are technically simple and thus can be enhanced with new ideas. In this work, we propose a new TSRL framework, namely multi-source multi-phase time series representation transfer (SMart), which has two novel mechanisms to address the aforementioned deficiencies: 1) a multi-phase recurrence plots recovery task, in three alternative modes, for guiding the encoder to embed time series dynamics into the time series representation; and 2) a source dataset selector to select multiple suitable source datasets to supplement the original target dataset for pre-training the TSRL encoder. Experimental results show that SMart outperforms several state-of-the-art models for time series representation learning, classification and regression on both uni-variate and multi-variate time series datasets, reducing mean absolute error up to 19.5% for time series regression, and increasing average accuracy up to 1.34% for time series classification.
Read the full abstract

Quantum Algebra

The tensor category for W(2,2)-vertex algebra

This paper studies the category $\mathcal{C}$ of grading-restricted $C_1$-cofinite generalized modules for the vertex operator algebra associated to the $W$-algebra $W(2,2)$. We first show that $\mathcal{C}$ is the same as the category of finite length modules whose simple composition factors are the irreducible highest weight $W(2,2)$–modules $L[r]$ of highest weight $(\frac{1-r}{2}, 0)$ for $r \in \mathbb{Z}_{> 0}$, and hence $\mathcal{C}$ carries a braided tensor category structure. Then we prove the fusion rules for the simple objects $L[r]$ are governed by the $\mathfrak{sl}_2$ Clebsch–Gordan rule. In particular, we prove \[ L[r]\boxtimes L[s]\cong \bigoplus_{i=0}^{\min\{r,s\}-1} L[r+s-1-2i]. \] Using the fusion rules and a recent result of Etingof–Penneys, we establish the rigidity of $\mathcal{C}$. We also show the semisimple subcategory generated by the simple objects $L[r]$ is tensor equivalent to the category Rep $\mathfrak{sl}_2$ of finite dimensional $\mathfrak{sl}_2$-modules.
Read the full abstract

All authors Drazen Adamovic, Mingjie Peng, Gordan Radobolja, Jinwei Yang

Biological Physics

Trypanosome motility in blood and particle suspensions

Motile microorganisms often navigate crowded and structurally complex environments, where surrounding obstacles can strongly influence locomotion. The bloodstream form of the flagellate parasite Trypanosoma brucei circulates in blood, a dense suspension of red blood cells (RBCs), yet the physical mechanisms governing its locomotion under such conditions remain poorly understood because direct experimental observations are challenging. Here, we combine numerical simulations with in vitro experiments to investigate trypanosome motility in concentrated RBC suspensions and in suspensions of spherical colloidal particles. Simulations reveal that the parasite swimming speed increases by up to $50\%$ at RBC volume fractions comparable to those in blood. To identify the origin of this enhancement, we perform controlled studies in colloidal suspensions with particles of different sizes. We find that suspended particles substantially increase the anisotropy between the perpendicular and parallel friction coefficients acting on the beating flagellum, thereby enhancing propulsion. This effect is most pronounced when the suspended particles are comparable to or smaller than the characteristic wavelength of the flagellar beat. Experiments with microparticle suspensions confirm an increase in trypanosome propulsion with increasing particle concentration, in qualitative agreement with the simulations. Our results uncover a general physical mechanism by which concentrated particle suspensions can enhance flagellar-beat-driven locomotion and suggest that the densely crowded, particulate environment of blood may facilitate trypanosome propulsion. These findings provide new insight into trypanosome motility in the bloodstream and may apply broadly to other flagellated microswimmers in complex suspensions.
Read the full abstract

All authors Florian A. Overberg, Marie Kater, Timothy Krüger, Gerhard Gompper, Markus Engstler, Dmitry A. Fedosov

Quantum Physics

Observation of Hong-Ou-Mandel interference between photon and polariton

Light-matter interactions underlie many quantum technologies, yet whether quasiparticles formed from such interactions preserve the full quantum state of light remains unresolved. Surface plasmon polaritons (SPPs), a class of polaritons formed by interacting photons with free-electron oscillations at metal-dielectric interfaces, are prime candidates to explore this question. Here we demonstrate quantum interference between single photons and SPPs using an Au-SiN$_{\mathrm{x}}$ integrated photonic-plasmonic device. Our results reveal that SPPs retain the indistinguishability of their excitation photons, establishing SPP as a viable quantum information carrier and opening a potential route toward photonic-plasmonic quantum circuitry.
Read the full abstract

All authors Yun-Ru Fan, Ying-Ao Su, Kai Guo, Bo-Yu Fan, Yao-Qing Zhang, Hai-Zhi Song, Hao Li, Yong Geng, Kun Chen, Deng-Ke Zhang, Li-Xing You, Yan-Yu Wei, Guang-Can Guo, Qiang Zhou

Materials Science

The Locality Cost of Fully Flat Hopf Insulators

Hopf topology permits a strictly finite-range Hamiltonian with one exactly flat topological band. We prove, however, that extending flatness to the complete two-band spectrum necessarily sacrifices strict locality or the gap: any gapped, Hermitian, translationally invariant two-band Hamiltonian with strictly finite-range hopping and two exactly flat bands has vanishing Hopf invariant. Equivalently, within this two-band setting, a Hopf band admits no compactly supported, translation-covariant, orthonormal Wannier generator. For factorized one-flat-band Hopf parents, the unavoidable partner dispersion equals the Gram symbol of translated compact localized states and encodes their nonorthogonality. Full flattening converts this dispersion into exponentially decaying but infinitely supported hopping. Model-independent bounds provide a sufficient criterion for finite-range approximants to retain the Hopf phase. An explicit model yields the axial decay length $\xi_z/a=1/\ln 2$, parameter-free hopping tails, and residual bandwidths testable in circuit and photonic lattices. Hopf topology therefore does not prohibit a flat band but forces the locality–flatness cost to appear as either partner-band dispersion or nonlocal hopping.
Read the full abstract