Combinatorics
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Monday 14 September
Let $\sigma_{n,d}$ denote the elementary symmetric Boolean function of $n$ variables and degree $d$. Previous work determined $C(\sigma_{n,d})$ when $d$ is odd or a power of two, but the general even non-power-of-two case remained open. We determine the certificate complexity for every degree $1\le d\le n$, thereby completing the classification for elementary symmetric Boolean functions. Writing $d=2^t m$ with $m$ odd, we obtain an explicit formula in which the possible deficit from the maximal value $n$ is controlled by $2^t$, while the exact value is determined by a binary containment condition involving $m$. In particular, \[ n-2^{\nu_2(d)}+1\le C(\sigma_{n,d})\le n, \] and we characterize when the upper bound is attained. Moreover, we determine the least positive period of the certificate-complexity deficit $\Delta_d(n)=n-C(\sigma_{n,d})$: it is $1$ for odd $d$, equals $d$ when $d$ is a power of two, and equals $2^{\lfloor\log_2 d\rfloor+1}$ for even non-power-of-two $d$. This least-period problem is distinct from the classical periodicity of the underlying value sequence $\binom{j}{d}\bmod 2$. The known odd-degree and power-of-two formulas are recovered as special cases.
Data Structures and Algorithms
Breaking the 1/3 Barrier for constrained $\boldsymbol{k}$-Submodular Maximization: A Proportional Top-2 Randomized Framework
$k$-submodularity generalizes submodularity by allowing each selected element to be assigned one of $k$ labels. We study the maximization of a non-monotone $k$-submodular function, where $k\ge2$, under matroid and knapsack constraints. We develop a proportional top-2 randomized framework that achieves an approximation ratio of $\sqrt2-1\approx0.4142$ under either constraint, breaking the previous $1/3$ barrier. Once an element is selected, its label is chosen from the two labels with the largest marginal gains, with probabilities proportional to their positive parts. The analysis separates replacement, pure-addition, and residual-deletion losses, and proves a sharp local loss certificate for this rule. We then give a Bellman analysis using the same local certificate. It gives a $1/2$-approximation under a cardinality constraint and strengthens the knapsack guarantee to $1/2$ without changing the algorithm. The $1/2$ guarantees are oracle-optimal for $k=2$ and asymptotically tight as $k\to\infty$. The value-oracle complexities are $O(Bnk)$ for cardinality, where $B$ is the cardinality bound, $O(n^2k)$ for matroid, and $O(n^3k^2)$ for knapsack.
Read the full abstract
Computer Vision
MambaMPD: A Mamba-Driven Segmentation Framework for Marine Pollution Detection from Remote Sensing Imagery
Accurate marine pollution detection (MPD) is essential for protecting coastal ecosystems and marine biodiversity. Vision Mamba models have shown promise in remote-sensing semantic segmentation by efficiently capturing long-range dependencies and global context, yet their potential for MPD remains underexplored. MPD is particularly challenging because of low signal-to-noise ratios, fragmented pollution patterns, and indistinct boundaries caused by the visual similarity between pollutants and the surrounding sea. To address these issues, we propose MambaMPD, an enhanced Mamba-based framework incorporating two complementary structural priors: Frequency-Aware Augmentation (FAA) and multi-scale Edge-Guided Attention (EGA). FAA integrates wavelet transforms into the encoder to decompose features into multi-scale frequency subbands, enabling the model to capture low-frequency contextual semantics and high-frequency structural details needed to identify small, low-contrast, and irregular pollution patterns. EGA adaptively fuses hierarchical, Laplacian-derived boundary cues with deep semantic representations, refining encoder features before decoding to sharpen boundaries and reduce ambiguity in visually confusing, spatially fragmented scenes. Together, these modules improve sensitivity to subtle pollution signals while preserving fine boundary structures. A U-Net-style decoder with squeeze-and-excitation attention and deep supervision progressively restores and refines semantic and spatial information across scales. Extensive experiments on two benchmark MPD datasets show that MambaMPD achieves higher mIoU than competing methods while requiring substantially less computation than foundation-model-based approaches. On MADOS, it improves F1 by 3.6% over OSDMamba; on M4D, it raises Oil Spill IoU by 6.82% over TransOilSeg.
Read the full abstract
Sound
Sectional Structure and Emotional Dynamics in Chinese Pop Songs: An Empirical Analysis of Valence-Arousal Trajectories across 100 Songs
Music Emotion Recognition (MER) aims to identify and represent emotional information in music through computational methods and is an important research area within Music Information Retrieval (MIR). To address the limited consideration of song sectional structure in existing dynamic MER research, this study examines 100 Chinese pop songs by aligning 1,046 manually annotated sections with continuous Valence-Arousal (VA) trajectories and analyzing them from the perspectives of section type, adjacent section transitions, repeated sections, and whole-song trajectories. The results show that emotional differences across sections are reflected more strongly in Arousal. Verse typically forms a relatively low-activation baseline, Pre-chorus exhibits a progressive buildup, and Chorus produces a more pronounced high-arousal arrival, while Interlude, Bridge, and Outro tend to show transitional, divergent, and closing functions, respectively. Although whole-song sectional configurations are diverse, high-frequency local transitions are relatively concentrated. High-arousal positions occur more often in the later part of a song, but are not fixed to the final Chorus. Based on these findings, this study summarizes the emotional organization of the selected pop songs as an empirical framework of "local cycles-global accumulation", in which local sectional cycles are accompanied by emotional pullbacks, while the overall trajectory exhibits a later-stage rise in VA and a tendency for high-Arousal positions to occur later in the song.
Read the full abstract
Mesoscale and Nanoscale Physics
Opto-Electrical Detection of Donor Bound Excitons in Silicon-on-Insulator Substrate
Spin of a donor bound electron in silicon has been shown to be a very coherent qubit system but lacks a coherent optical interface. There does, however, exist a donor bound exciton transition that can be excited optically but decays dominantly via an Auger recombination producing an electrical signal. This provides an opto-electrical pathway for spin readout. Scaling this readout to single-spin level will require interfacing the spins with silicon photonics for efficient guiding of photons, which in turn will require moving to silicon-on-insulator (SOI) substrates. Here we demonstrate first ensemble opto-electronic measurements of donor bound excitons in SOI material, including isotopically purified 28-Si device layers. The experiments show pronounced shifts in the resonance wavelengths and broadenings of the transition linewidths compared to the bulk experiments.
Read the full abstract
Differential Geometry
Polarization complete left invariant connections
Let $(M,\nabla)$ be a real analytic Koszul manifold. An adapted complex structure (ac-structure) on a neighborhood $N$ of the zero section in $TM$ is a complex structure on $N$ such that the leaves of the Levi-Civita foliation are holomorphic curves. More generally, a complex polarization $P$ on $N$ is called an ac-polarization if the leaves of the Levi-Civita foliation are tangential to $P$. The bundle of (1,0) tangent vectors of an ac-structure is an ac-polarization. The connection is called entire (resp. polarization complete or simply $\mathcal P$-complete) if the ac-structure (resp. the ac-polarization) exists on $TM$. Although on a small enough $N$ an ac-structure always exists, entire connections are rear and if a maximal domain of definition $N_{max}$ of an ac-structure exists (different from $TM$), $N_{max}$ is a complicated domain. On the other hand in many cases the associated ac-polarization can be extended to the whole $TM$. The main purpose of the paper is to gain better understanding of this phenomenon using a generalized version of the polar map. As special cases we show that many of those metrics studied by Aslam-Burns-Irvine and Halverscheid-Iannuzzi although are not entire but are $\mathcal P$-complete.
Read the full abstract
High Energy Astrophysical Phenomena
Compelling evidence of a link between the lags of the quasi-periodic oscillations and the radio jet in the black-hole X-ray binary GRS 1915+105
GRS 1915+105 is one of the most studied black-hole X-ray binaries, known for its extreme variability and rich phenomenology. Previous studies of this source with the Rossi X-ray Timing Explorer reported a transition of the phase-lags of type-C quasi-periodic oscillations (QPOs) from soft, where low-energy photons lag the high-energy ones, when the QPO frequency is higher than ${\sim} 2 ~{\rm Hz}$, to hard when the QPO frequency decreases below ${\sim} 2 ~{\rm Hz}$. The hard-lags of the QPO coincide with episodes of strong radio emission. We analyse NICER observations of GRS 1915+105 obtained between 2018 and 2020, during a period in which the source flux decreased steadily, and perform a detailed spectral-timing study of the detected type-C QPOs. We find a type-C QPO with frequencies in the range of ${\sim} 1.3$-$3.9 ~{\rm Hz}$, which displays soft lags and, contrary to the RXTE observations, shows no evidence of hard phase lags at frequencies below $2~{\rm Hz}$. Quasi-simultaneous AMI-LA radio observations show consistently low radio flux (${\lesssim} 5 ~{\rm mJy}$) during this period. These results appear to show that the hard QPO lags in GRS 1915+105 are linked to the presence of strong radio activity, suggesting that the relativistic jet is responsible for the hard phase lags, supporting a scenario in which QPO phase lags trace changes in coronal geometry and accretion-ejection coupling in GRS 1915+105.
Read the full abstract
Computer Vision
Don't Send What You Don't Need: Question-Guided Token Pruning as a Privacy Defense for Vision-Language Models
Visual Question Answering (VQA) with Vision-Language Models (VLMs) is increasingly used in privacy-sensitive and bandwidth-constrained settings. Federated Learning (FL), Split Learning (SL), and U-Shaped Split Learning (USL) keep raw data local, but transmitting all visual tokens across a model partition remains costly and can expose private information. We propose QPriv-VL, a question-guided, privacy-aware token-pruning framework for FL, SL, and USL that prunes visual tokens before transmission based on task utility and privacy sensitivity. Its core component is a lightweight Dynamic Threshold Predictor (DTP) that jointly estimates a sample-specific pruning ratio and a token-level retention mask in one forward pass. DTP combines question relevance, computed from cross-modal similarity between visual patches and the pooled question embedding, with a sensitivity signal derived from frozen DINOv2 features. This allows the model to suppress potentially sensitive regions while preserving patches useful for answering the question, without requiring sensitivity labels. We evaluate QPriv-VL on GQA, OK-VQA, VQAv2, SLAKE, VQA-RAD, and PathVQA against four privacy attack families: FSHA, FORA, iDLG, and attribute-inference membership inference attacks. DTP matches or outperforms fixed-ratio pruning while using substantially fewer transmitted tokens. On VQA-RAD, it reduces membership-inference attack success from 0.99 to 0.76-0.79, lowers FSHA and FORA reconstruction PSNR relative to fixed-ratio pruning, and preserves competitive VQA accuracy using about 40% of the original visual-token budget. A sensitivity exclusion ratio of 1.20 +/- 0.18 indicates preferential removal of privacy-sensitive patches, while explainability analysis shows that retention adapts to question semantics rather than generic visual saliency.
Read the full abstract
Rings and Algebras
The Cuntz semigroup of rings with stable rank one
We study the Cuntz semigroup of arbitrary rings with stable rank one. We prove that Cuntz subequivalence between countably generated projective right modules $P$, $Q$ is equivalent to $P$ being isomorphic to a pure submodule of $Q$, whilst Cuntz equivalence amounts to isomorphism.~We also prove order-cancellation of finitely generated projective modules (and, more generally, pairs of modules admitting certain complements) from direct sums with countably generated projective modules.
Read the full abstract
Computer Vision
Benchmarking Intra-Patient 3D Deformable Multimodal Image Registration
Multimodal image registration is a key component of many clinical workflows, yet it remains challenging because corresponding anatomical structures often exhibit substantially different image intensities across modalities. In this work, we present a comprehensive benchmark of intra-patient 3D multimodal deformable registration methods across three datasets covering different anatomical regions and difficulty levels, including both synthetic deformation recovery and real clinical scenarios. We evaluate classical optimization-based approaches and modern learning-based methods, including recent deep learning and foundation models, using complementary metrics: Average Dice similarity coefficient (DSC), average 95th-percentile Hausdorff distance (HD95), and a modality-independent structural similarity measure based on the MIND self-similarity context (MIND-SSC). Results show high variability across datasets, with learning-based methods demonstrating superior performance on large synthetic benchmarks, while only limited improvements are observed in real pelvic registration. A key finding of this study is the consistent disagreement between geometric metrics (DSC, HD95) and image-based similarity metrics (MIND-SSC), highlighting that improved overlap does not necessarily imply better global multimodal correspondence. Furthermore, anatomy-guided approaches achieve the highest overlap scores but exhibit degraded performance outside of segmented regions, revealing a trade-off between label-driven alignment and global structural coherence. Overall, our results indicate that no current method achieves robust performance across anatomies and modalities. We demonstrate that intra-patient 3D multimodal registration requires multi-criteria evaluation, including deformation-based metrics, and remains an open problem.
Read the full abstract
Computer Vision
Circuit-MLLM: Topological Logic-Guided Latent-Space Visual Reasoning for Circuit Schematic Understanding
Through pre-training on extensive text and image datasets, current multi-modal large language models (MLLMs) achieve strong performance on general tasks. However, circuit schematics present a unique challenge for MLLMs due to their dense component layouts and distinct topological logic, demanding fine-grained structural parsing to extract the electrical semantics. To address this, we propose Circuit-MLLM, a multimodal reasoning framework that reformulates circuit topology analysis as a process of device localization, path tracing, and sequential reasoning within the latent space. We introduce a circuit knowledge mining mechanism that deeply aligns the model's latent representations with structurally rich features derived from multi-granularity circuit vision experts, enabling the model to effectively internalize topological semantics. Building upon these internalized semantics, we devise a topology-guided sequencing strategy that decouples reasoning from the rigid raster-scan order, enforcing stepwise inference along the circuit's topological logic in latent space. Across diverse circuit analysis tasks, Circuit-MLLM consistently outperforms strong baselines, notably achieving a 25% higher average score than GPT-5.1, which demonstrates the effectiveness of our framework in circuit schematic topology analysis. Code is publicly available at https://github.com/IC-Yuan/Circuit-MLLM.
Read the full abstract
Robotics
Tracking the Ground: Online Lidar Identification of Robot-Induced Soil Deformation in Agricultural Environments
Agriculture faces many challenges, and robotic systems can play an important role in addressing them by improving the efficiency and sustainability of field operations. Among these challenges, preserving soil health is a critical concern, as vehicle-soil interactions can degrade the soil structure and produce unwanted surface deformation. A key step toward soil-aware robotics is to explicitly account for how vehicle traffic deforms the ground, yet soil state is typically not treated as a variable. We address this gap by proposing a framework to quantify traffic-induced soil deformation and estimate its evolution online from lidar observations. The method relies on a reduced-order parametric model that represents the soil behavior via physically interpretable parameters, yielding a continuously updated and observable representation of soil state. Experiments conducted in different soil conditions demonstrate the ability of the approach to capture deformation induced by the robot. By making soil response measurable and interpretable during operation, the proposed framework establishes a basis for soil-aware robotic operation, in which the estimated state can be exploited to adapt robotic behaviors in order to reduce soil degradation.
Read the full abstract
Audio and Speech Processing
OpenEnded: An Open-Response Speech Corpus for Speaking Proficiency Assessment with Human Annotations and ALM Supervision
The development of automated speaking assessment (ASA) is limited by the scarcity of public datasets, with most existing work relying on read-aloud speech, which limits applicability to real-world communication scenarios. In this work, we introduce OpenEnded, a corpus of English practice speech from Mandarin speakers in open-response tasks. Unlike prior open-response datasets that provide only holistic proficiency scores, OpenEnded offers utterance-level assessments of accuracy, fluency, and prosody. Approximately 10,000 utterances are collected and annotated using a hybrid framework: 1,000 are manually labeled via multi-rater scoring with discrepancy resolution to form a high-quality test set, while the remaining are pseudo-labeled by an audio language model (ALM) for training and development sets. We evaluate ALMs and existing ASA models on the OpenEnded test set and introduce VoxPA as an additional baseline. Results show that ALM-generated pseudo-labels improve training over original ALM scoring, while VoxPA achieves the best performance among all baselines.
Read the full abstract
Combinatorics
Algebraic Structures on Sets of Partitions
Motivated by Robert Schneider's trailblazing work toward developing a unifying algebraic theory of integer partitions, we explore various binary operations on partitions to identify algebraic structures on sets of partitions. In particular, we discover that several sets of restricted partitions form abelian groups under reduced versions of concatenation, component-wise addition, and component-wise multiplication. One type of restricted partition from a group structure also enjoys a bijection with ordinary partitions of any given size. We extend two partition groups to vector spaces over the finite field $\mathbb{Z}_p$, where $p$ is a prime. We further discover that partitions are equipped with a commutative ring structure. Finally, we consider subgroups, subspaces, and ideals of our algebraic partition structures to investigate properties of related types of restricted partitions. The new examples of algebraic structures described in this paper open the door to partition analysis via algebraic tools, decompositions, extensions, and geometry.
Read the full abstract
Distributed Computing
Cloud-Edge LLM Collaborative Learning via Accelerating Personalized Bilevel Optimization
Amid the rapid advancement of physical-world intelligence, cloud-edge collaborative large language models (LLMs) have emerged as a promising roadmap for practical LLM deployment. However, existing cloud-edge paradigms struggle to balance global consensus with local personalization, which fails to satisfy the need for a unified knowledge foundation on the cloud and domain-specific adaptation at the edge. To address this, we propose a personalized bilevel optimization framework that formalizes cloud-edge LLM collaboration as a dual structure: the upper level optimizes edge-side personalization, while the lower level governs cloud-side knowledge transfer, reaching cloud-edge evolving in coordination. We then propose CIDERS: while the cloud performs knowledge transfer to the learnable backbone, the key lies in embedding global trajectories into each local personalization step via consensus-variate correction to reconcile personalization with consensus. We provide a comprehensive theoretical analysis, including a geometric characterization of the local trajectory and a full convergence guarantee, revealing an explicit trade-off structure between personalization and global convergence. Extensive experiments on Qwen2.5-3B and Qwen2-1.5B show CIDERS high performance and can significantly improve the accuracy (as much as 200%+). Control-state diagnostics further show strong preservation of the local orthogonal component, while a separate cross-domain evaluation finds only a small matched-domain response advantage.
Read the full abstract
Superconductivity
Thermal Hall effect in elemental niobium, a two-gap superconductor
Niobium holds a pivotal place in superconductivity history: it is not only the elemental superconductor with the highest critical temperature, but also a long-standing candidate for multiband pairing whose evidence has remained controversial. A two-gap scenario was proposed as early as 1959, yet experimental proof stayed elusive. Here, through thermal Hall effect measurements, we unambiguously demonstrate a crossover in transverse thermal transport below $T_{\rm c}$, where the dominant carrier switches from hole-like to electron-like upon cooling. This crossover is a clear hallmark of two distinct superconducting condensates. Fitting our data to a two-gap Bardeen-Rickayzen-Tewordt (BRT) model yields a second energy gap of approximately 0.22 $k_{\rm B}T_{\rm c}$, only 11% of the dominant gap. This small gap size accounts for the difficulty in resolving the two-gap structure in earlier experiments. Crucially, the temperature at which the electron-like contribution to longitudinal thermal conductivity begins to dominate coincides with the onset of the sign reversal in the thermal Hall coefficient, confirming consistency between the two methods, and providing crucial information on the assignment of the superconducting gaps to their respective hole- or electron-like Fermi surface sheets. These findings not only resolve a longstanding controversy, but also establish that multigap superconductivity is far more common than previously assumed, and demonstrate the thermal Hall effect as a powerful probe for resolving gap multiplicities in superconductors.
Read the full abstract
Quantum Physics
Efficient quantum state tomography with two complementary projective measurements
Quantum state tomography (QST) is of fundamental importance to characterize quantum systems in quantum information processing, but its practical implementation is severely hindered by the exponential scaling of measurement and computational costs. In this paper, we present a novel QST protocol that utilizes Kirkwood-Dirac (KD) quasiprobability to reconstruct quantum states. First, it enables state reconstruction with only two complementary rank-one projective measurements, thus significantly reducing the measurement cost. Then, a complex logistic regression estimator is proposed to process collected KD data, together with a projected gradient algorithm to mitigate numerical instability and to accelerate convergence. The product-operator structure of KD quasiprobability is further exploited to reduce the computational cost. Finally, extensive experiments are implemented to confirm the validity of our protocol. Notably, the full reconstruction of randomly generated 15-qubit mixed-state instances can be accomplished within 20 minutes under the GPU implementation. These results suggest a promising route toward scalable QST and benchmarking large-scale quantum systems.
Read the full abstract
Optics
Extracting dipole orientations from asymmetric plasmonic nanostructures towards machine-learning-assisted spectropolarimetry
In this work, nanoparticles with various asymmetries are analyzed for their azimuthal orientations using polarimetric dark-field spectroscopy at different analyzing angles of a linear polarizer. This approach reveals their spectral behavior in terms of electric far-field dipole intensities when modeled with an analytical dipole model. By simultaneously fitting the spectra from a set of analyzer angles, the respective dipole orientations are extracted. In a statistical approach, all non-repeating permutations are further studied with a machine learning algorithm. The resulting azimuthal distribution of dipole orientations coincides well with the geometric orientations derived from simulations and electron microscope images. A histogram gradient boosting regressor evaluates the impact of the measurement setup on the simultaneously fitted sets, linking the weights of the analyzer angles to the asymmetry in the plasmonic systems. This comprehensive spectroscopic method improves the accuracy of dipole orientation measurements and enables modern machine learning models to interpret potentially complex features of nanostructures.
Read the full abstract
Cryptography and Security
CiteShade: Citation Laundering in Multi-Source Retrieval-Augmented Generation and Its Counterfactual Defense
Retrieval-augmented generation (RAG) grounds a language model's answers on retrieved external knowledge and returns each answer with citations that identify its sources. Those citations are the user's audit trail: they let a reader verify a claim without trusting the model. Prior security work on RAG asks whether an attacker can corrupt the answer, leaving the citation channel unexplored. We show that this channel is a new and practical attack surface. We propose CiteShade, the first citation laundering attack to RAG, in which an attacker controlling a single source induces a model to produce an attacker-chosen wrong answer and to attribute it to a trusted source that does not support it, while the evidence for the correct answer remains in context. We formulate the attack as an optimization problem, derive three necessary conditions (retrieval, generation, and citation) and construct sources satisfying them without any instruction. On multi-source multi-hop question answering the attack raises the wrong-answer rate from 0.01 to 0.68, and source deletion confirms the malicious source is the causal driver in every measured case. Vulnerability tracks a model's propensity to cite rather than its scale, reaching CLR 0.84 under explicit instruction and 0.64 with no instruction at all on the most citation-prone model tested. We then show that perplexity filtering and citation-support checking are each insufficient, and propose a counterfactual defense that verifies which source actually drove the answer.
Read the full abstract
Graphics
KaiNinja: Extending Native 3D Generators to the Part Level
Native 3D generators turn one image into a single mesh. TRELLIS.2 and its peers deliver high-fidelity non-watertight geometry with materials, but the output is one fused object, while downstream work such as editing, rigging and simulation operates on part-level assets. A naive idea is to run a 3D segmentation network on the fused mesh that TRELLIS.2 generates, but such pipelines are slow and bounded by the accuracy of the segmentation. We want a simple way to extend an existing native 3D generator to the part level. But we face a critical problem: the O-Voxel grid stores one sheet of surface per voxel, so a single volume cannot represent the interface where two parts touch, at any resolution. We introduce a dual-volume representation to solve this problem and put forward KaiNinja, a part-level extension of TRELLIS.2 built on a dual-volume form of its O-Voxel representation. KaiNinja keeps the generation speed and quality of TRELLIS.2 while extending it to the part level, with no mask or segmenter in the pipeline. Its training data come from sources of many kinds, including CAD models and assets authored by an LLM-driven agent; to our knowledge it is the first 3D generative model trained on agent-authored part data. Surprisingly, we also find that whole-object fidelity improves over the same backbone fine-tuned on the same dataset. Against part generation pipelines of different paradigms, it lowers whole-object Chamfer distance by 40% and raises strict part F-score by 16%.
Read the full abstract
Number Theory
Singular automorphic products and BKM algebras
After proving the moonshine conjecture, Borcherds asked in 1995 whether there exist only finitely many holomorphic automorphic products of singular weight. We show that, up to conjugation, there are exactly 27 such products on lattices of the form $2U\oplus L$. The classification proceeds by identifying each product at a standard $1$-dimensional cusp with the superdenominator of a BKM superalgebra that carries a graded module structure over an affine Lie superalgebra $\hat{\mathfrak{g}}$. The auxiliary algebras $\hat{\mathfrak{g}}$ are shown to satisfy strong conditions that reduce the possibilities to an initial list of 1372 candidates. An elimination argument leaves exactly 85 algebras $\hat{\mathfrak{g}}$ that correspond to the 27 products. This classification is closely connected with certain distinguished families of vertex operator superalgebras. Our results provide a systematic framework for extending affine Lie superalgebras to hyperbolic ones with modularity, generalizing the work of Feingold and Frenkel (1983).
Read the full abstract
Machine Learning (Statistics)
Predictive Likelihood Ratios for Language Model Watermark Detection
Keyed watermark detection tests dependence between observed tokens and pseudorandom variables reconstructed from a secret key. Building on the pivotal framework of Li et al. (2025), we construct predictive likelihood ratios that average over uncertain probability deficits and residual-tail distributions. The aim is robust detection power across alternative specifications without requiring a single signal-strength tuning. A mixture prior combines tail shape and effective width; hierarchical extensions allow within-document variation in deficit or width. The test maximizes prior-averaged power at a fixed size, but is not generally uniformly most powerful or minimax. Under the exact conditional pivot null, normalized predictive alternatives selected before each observation yield a Bayes factor that is also a test martingale: Type I error control is unaffected by alternative misspecification and remains valid under optional stopping. This guarantee does not cover violations of the conditional null, and the interpolated implementation has no certified anytime guarantee. Gumbel marginal likelihoods are evaluated by fixed quadrature. Across the evaluated tail-shape and tail-width alternatives and three horizons, the union-tail mixture has maximum observed Type II error regret .0080, compared with .0962 for the equal-tail mixture, relative to the best tested rule. On temperature-matched outputs from two open models, it improves AUC over the equal-tail baseline in all eight non-saturated model-temperature cells, although the leading reference score generally has higher AUC. Supplementary experiments show retained power under independent null-like replacement and smaller changes from hierarchical dependence modeling. The evidence supports robustness across the evaluated alternatives, not uniform power guarantees or resistance to arbitrary text edits.
Read the full abstract
High Energy Physics (Phenomenology)
Hyperon polarization in deep inelastic scattering
We have implemented the recent extension of the string+${}^3P_0$ model of spin-dependent hadronization that includes the production of spin-$1/2$ baryons in the Pythia event generator for the simulation of deep inelastic scattering (DIS). The implementation in Pythia is performed by extending the StringSpinner package, which is then used to study the spontaneous polarization of $\Lambda$ and $\bar{\Lambda}$ hyperons produced in the current and target fragmentation regions in DIS with an unpolarized proton target. We find a large spontaneous polarization for $\Lambda$s in the target fragmentation region, calling for the measurement of such observable that can shed light on the quark spin dependence of hadronization.
Read the full abstract
High Energy Physics (Experiment)
First Observation and Dynamical Study of the $D^+_s\to f_{0}(980) \mu^+\nu_\mu$ Decay
Using 7.33 fb$^{-1}$ of $e^+e^-$ annihilation data recorded with the BESIII detector at center-of-mass energies from 4.128 to 4.226 GeV, we report the first observation and dynamical study of the semileptonic decay $D^+_s\to f_{0}(980) \mu^+\nu_\mu$. The absolute branching fraction of $D^+_s\to f_{0}(980) \mu^+\nu_\mu$ with $ f_{0}(980)\to \pi^+ \pi^-$ is $(1.59 \pm 0.18_{\rm stat} \pm 0.11_{\rm syst}) \times10^{-3}$. Combining this result with our earlier BESIII measurement of ${\mathcal B}(D^+_s\to f_{0}(980) e^+\nu_e)$, their ratio is found to be $\frac{{\mathcal B}(D^+_s\to f_{0}(980) \mu^+\nu_\mu)}{{\mathcal B}(D^+_s\to f_{0}(980)e^+\nu_e)} = 0.92\pm0.13_{\rm stat}\pm0.08_{\rm syst}$, in agreement with the Standard Model expectation of lepton flavor universality. From a dynamical analysis of the $D_{s}^{+} \to f_{0}(980)\mu^+\nu_\mu$ decay with a simple pole parametrization for the hadronic transition form factor, the product of the form factor $f^{f_{0}(980)}_{+}(0)$ and the $c\to s$ Cabibbo-Kobayashi-Maskawa matrix element $|V_{cs}|$ is determined to be $f^{f_{0}(980)}_{+}(0)|V_{cs}|=0.490\pm0.059_{\rm stat}\pm0.025_{\rm syst}$. Averaging with our previously reported result for the $D_{s}^{+} \to f_{0}(980)e^+\nu_e$ decay, we obtain $f^{f_{0}(980)}_{+}(0)|V_{cs}|=0.500\pm0.016_{\rm stat}\pm0.020_{\rm syst}$. Using $|V_{cs}|$ from the CKMfitter group, we extract $f^{f_{0}(980)}_{+}(0)=0.514\pm0.017_{\rm stat}\pm0.021_{\rm syst}$. This represents the most precise determination of the $D_{s} \to f_{0}(980)$ transition form factor to date, and provides stringent tests of various theoretical models.
Read the full abstract
Computation and Language
Empathy Is Steerable but Multi-Axial: Mechanism Geometry and Persona Effects in LLMs
Activation steering has been used to control traits such as honesty, refusal, and sycophancy, yet supportive empathy is evaluated along multiple dimensions that need not correspond to independently controllable activation directions. Using the EPITOME framework, which decomposes supportive empathy into Emotional Reactions, Interpretations, and Explorations, we study three instruction-tuned LLMs and ask whether candidate directions derived from these labels produce distinguishable intervention effects or instead share structure, and how persona prompts interact with those directions. We find that contrastive activation addition yields a stable middle-layer intervention that consistently shifts the EPITOME proxy scores across models, moving empathy analysis beyond response-level scoring. However, the recovered directions are only partially separable: steering one direction induces off-target shifts, and hand-crafted prompting shifts the empathy profile rather than isolating a single dimension. Persona prompts substantially change EPITOME scores, but a paired activation-shift decomposition shows that the recovered subspace captures only approximately 3 percent of persona-induced squared activation-shift magnitude at layer 15. Under this EPITOME-based definition, expressed empathy is steerable but multi-axial, and controlling persona-conditioned empathy requires targeting structure beyond individual mechanism directions.
Read the full abstract
Category Theory
Mixing Extriangulated Model Structures
Let $(\mathscr{C}, \mathbb{E}, \mathfrak{s})$ be a weakly idempotent complete extriangulated category, and let $\xi_1\subseteq\xi_2$ be proper classes of $\mathbb{E}$-triangles. We prove an analogue of Cole's mixing theorem in this setting: two compatible admissible model structures, relative to $\xi_1$ and $\xi_2$ respectively, give rise to a mixed admissible model structure $\mathcal{M}_m$ relative to $\xi_1$, which has the trivial objects of the second model structure and the fibrant (or cofibrant) objects of the first. We also describe the cofibrant objects of $\mathcal{M}_m$ explicitly in terms of the two original model structures. For exact categories this recovers the mixing theorem for exact model structures, and for triangulated categories it applies to proper classes of triangles; we give a non-degenerate example on the homotopy category of a quasi-Frobenius ring.
Read the full abstract
Solar and Stellar Astrophysics
Stellar activity in a post-merger Giant star
Binary stars are the progenitors of exotic objects, such as supernovae and gravitational wave sources. However, some systems merge earlier when the primary expands into a giant star at the end of its lifetime on the main sequence and causes its companion to spiral in. TYC 4144-329-2 is a candidate post-merger first ascent giant star with a circumstellar disk seen in the infrared (IR). We detect weak X-ray and far ultraviolet (FUV) emission from TYC 4144-329-2, which we interpret as a signature of weak coronal activity. We also find variability in Ha line profiles and in the optical light curves and we suggest an intermediate geometry where the line-of-sight passes through the upper layers of a flared disk with a time variable column density. While the X-rays must be coronal, the Ha line profiles point to ongoing and variable accretion. We suggest that TYC 4144-329-2's merger was more recent than other stars in this class and that it did not yet have time to develop a deep convection zone. If this scenario holds, TYC 4144-329-2 would be a unique probe of the earliest stages in the evolution of post-merger systems.
Read the full abstract
Combinatorics
Multivariate stability of powered triangular recurrences and parametrised Eulerian polynomials
We prove multivariate stability for a class of affine triangular recurrences whose two coefficients are raised to an arbitrary positive integer power. The variables keep track of cross-step heights in a weighted lattice-path model. The applications include stability and real rootedness of parametrised Eulerian, Stirling, and Lah families. For the parametrised Eulerian polynomials, we obtain multivariate refinements of known real-rootedness results and prove simplicity and strict interlacing of consecutive rows. A kernel-preserving bijection with increasing binary trees yields a synchronised Foata–Strehl action on pairs and a combinatorial interpretation of their gamma coefficients along with a recurrence. The same bijection identifies refinement posets that admit symmetric Boolean and chain decompositions for parameter one and cover-supported $\mathfrak{sl}_2$ operators prove their PECKness. We extend the stability and strict interlacing results to synchronised Stirling permutations and prove PECKness for ordinary Stirling permutations with a fixed number of plateaux.
Read the full abstract
Sound
Graph Attention Design Choices Matter: A Controlled Study of LoRA-Adapted Audio Anti-Spoofing
Audio anti-spoofing systems increasingly combine self-supervised learning, parameter-efficient fine-tuning, and graph-attention-based backends. However, performance gains in such systems are often entangled with concurrent changes in the backbone, fine-tuning strategy, and training protocol, making the independent contribution of graph attention design difficult to isolate. To address this issue, we conduct a systematic controlled study of the graph attention layer under a unified experimental setting. We decompose the layer into three independently testable design dimensions: scoring symmetry, temperature learnability, and routing granularity. These are instantiated as a concat-based scoring branch, a LearnT branch with learnable temperature, and a multi-temperature routing branch, respectively. Each dimension is implemented as an independently gated residual branch, enabling the evaluation of both individual variants and their combinations under the same experimental setting. Experiments on five evaluation sets with five random seeds show that the LearnT branch achieves the best average equal error rate (EER), yielding a 16.1% relative improvement over the baseline. In contrast, the multi-temperature routing branch does not improve average performance on its own, but substantially reduces cross-seed standard deviation when combined with the concat-based scoring branch. Moreover, two individually effective branches degrade performance when used together, resulting in a 25.6% relative deterioration compared with the baseline. This finding reveals strong non-additive interactions among graph attention design dimensions. Overall, the results suggest that, under parameter-constrained fine-tuning, improvements in graph attention layers depend more on capacity allocation and branch interaction than on simply adding more learnable parameters.
Read the full abstract
Computer Vision
From Model Patterns to Abstract Semantics in Compositional Zero-Shot Learning
Compositional Zero Shot Learning aims to recognize unseen compositions by recombining learned primitives. Recent methods rely on vision language models and attempt to explicitly model contextual variations of primitives through multiple representations. However, such approaches are limited by fixed variant capacity and competition between abstract and concrete semantics. In this work, we present a new perspective that views primitive variations as the context-driven activation of concrete visual cues rather than independent entities. Based on it, we propose CLEAR, a CLoze-style rEAsoning-based Re-ranking framework inspired by human perceptual processes. CLEAR extracts conditional variants from the primitive candidate set in a coarse-to-fine manner, performs cloze-style reasoning to infer high-level semantics, and re-ranks predictions to correct biases toward salient concrete primitives. Extensive experiments demonstrate that CLEAR consistently improves the Base Model and outperforms state-of-the-art methods on the challenging C-GQA and MIT-States datasets. Code is available at https://github.com/buptLwz/CLEAR.