Peer Revenue Equity in science

Papers

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

All of arXiv

Sunday 30 August

Computer Vision

OPAL: Orthonormal Prototype Alignment Learning for Interpretable Image Classification

Prototypical part-based models provide explainable predictions by comparing input regions to learned prototypes. However, current approaches are burdened by complex, multi-stage training pipelines and heavily rely on auxiliary regularization to prevent prototype collapse. To overcome these limitations, we introduce Orthonormal Prototype Alignment Learning (OPAL), a single-stage, end-to-end framework that simplifies interpretable classification. Our approach anchors the latent space using predefined orthonormal bases, embedding each class within a dedicated subspace spanned by fixed part-prototypes. To achieve precise part localization, OPAL enforces spatial competition across feature maps. This mechanism isolates sparse, discriminative regions, directing each prototype to consistently attend to the same semantic concept across different images. By framing classification as a direct representation alignment task, our method eliminates the need for auxiliary losses. Extensive experiments on fine-grained benchmarks demonstrate that OPAL outperforms both its non-interpretable counterparts and state-of-the-art part-prototype methods, delivering granular visual explanations by explicitly revealing the specific image regions driving every prediction. Code is available at https://github.com/ilancarretero/OPAL.
Read the full abstract

All authors Ilán Carretero, Gustavo Jesús Angulo, Rocío del Amor, Valery Naranjo

Combinatorics

Avoiding patterns with three distinct letters in Canon permutations

We study avoidance of patterns of length $3$ with three distinct letters in canon permutations. We reduce the problem to studying pattern avoidance in lattice words and show that there are $6$ such pattern avoiding classes. This shows that there are $12$ classes for the original Canon permutation pattern avoidance problem. We also give descent refinements for these classes and classify the patterns for which the descent enumeration gives palindromic and $\gamma$-positive polynomials. When the polynomials are $\gamma$-positive, we explain the $\gamma$-positivity through a group action analogous to Foata-Strehl valley hopping. Additionally, we study the avoidance of patterns in the relabelling orbit of $1213, 12112, 1231$ after a conjecture about their cardinalities by Laudone and give bijective proofs for the results.
Read the full abstract

Superconductivity

Chemical Perspectives on Cuprate Superconductivity: Hole-Protected Spin Dimers as d Wave Cooper Pairs

The observation that the coherence lengths in cuprate superconductors are extremely small motivated us to search for its real-space structure by examining the consequence of hole-doping in CuO4 units in cuprates. We analyzed aggregates of hole-doped CuO4 units that phase-separate from hole-doped CuO2 layers in terms of their segregation-pressure indices defined in this work. Our search suggests that each d-wave Cooper pair is an antiferromagnetically coupled spin dimer protected in an antiferromagnetic hole dimer (AHD), and that the antiferromagnetic chains of AHDs provide channels through which such spin dimers can move collectively without perturbation. Based on these suggestions and their implications, we examined the causes for several seemingly puzzling observations on cuprate superconductors to find that they are naturally explained by the suggestions.
Read the full abstract

Networking

Matched-View Cross-Domain Evaluation of WireGuard VPN Traffic Classification Using Early-Flow Fingerprints

Classifying VPN-encrypted traffic by application category typically relies on datasets that collect non-VPN and VPN traffic in separate sessions, conflating encapsulation effects with session-level differences in user behavior, timing, and application mix. We use a recently published WireGuard tunnel dataset in which pre- and post-tunnel traffic is captured simultaneously, with a packet-level match ratio above 99.9%. This matched-capture design eliminates session-level confounds and enables a cross-domain benchmark: models are trained on non-VPN flows and tested on the VPN view of the same underlying flows. We compare whole-flow statistical aggregates (FlowFeatures) and Sequence of Packet Length and Time (SPLT) early-flow fingerprints across Random Forest, XGBoost, and a multi-scale CNN1D. Cross-domain transfer depends jointly on representation and model: tree ensembles achieve balanced accuracy of 0.84-0.93 with FlowFeatures but only 0.60-0.75 with flattened SPLT, whereas CNN1D processes the same SPLT fingerprint as a sequence and achieves the strongest transfer overall (balanced accuracy 0.98, macro F1 0.89) without any VPN data during training.
Read the full abstract

Sound

TEMPO: Temporally-grounded Multi-task Post-training for Large Audio-Language Models

Large audio-language models (LALMs) describe audio at the clip level but cannot assign timestamps to the events, speakers, or sounds they identify. Despite being essential for downstream tasks like speech recognition and dense audio captioning, timestamping remains a key limitation of most LALMs. We present TEMPO (Temporally-grounded Multi-task Post-training), the first unified model to handle audio, speech, and music timestamping tasks. Our core contribution is a supervised fine-tuning (SFT) stage built on three innovations: atomic timestamp tokens, a time-aware projector that injects sinusoidal wall-clock encodings into audio frame embeddings, and a distance-aware Gaussian loss. Our training is based on a synthetic-to-real curriculum. We further introduce, to our knowledge, the first application of reinforcement learning to unified audio timestamping, using GRPO with verifiable temporal rewards that directly optimize the evaluation objectives. Rather than serving as the primary source of performance gains, GRPO acts as a refinement stage on top of the SFT checkpoint, providing modest additional improvements. To support this work, we build a training dataset containing 119K samples and an evaluation benchmark containing 10K samples, drawn from established corpora across five tasks. On this benchmark, TEMPO outperforms Audio Flamingo Next and Qwen3-Omni, two state-of-the-art LALMs explicitly trained on timestamped data. Experiments confirm that SFT delivers most of these gains, with GRPO providing consistent but moderate refinements.
Read the full abstract

All authors Apoorva Kulkarni, Kaousheik Jayakumar, Sreyan Ghosh, Utathya Aich, Ramani Duraiswami, Dinesh Manocha

Machine Learning

The Intervention Gap in Latent World Models

Planning-time intervention fidelity is a distinct, measurable property of a learned world model: whether the model's own open-loop transitions move task variables the way matched environment interventions do. In the settings we test, it is neither revealed by reward fit nor ensured by task-anchored training. Across released TD-MPC2 checkpoint sizes, episode return falls as an operator-error diagnostic on task observables grows, while reward-prediction error stays small and nearly flat, and a self-supervised world model trained without task signal preserves the same operator substantially better than a task-anchored model on the shared task. A capture-gated matched-intervention audit then localizes what fails. On Cheetah, three LeWorldModel checkpoints capture the current task query and support decodable real intervention effects; however, their imagined five-step effects are worse than predicting no effect and worse than an environment-endpoint oracle. The failure is task-direction rotation with excess gain, not feature collapse. This severe pattern is conditional: five PreJEPA seeds retain an oracle-relative deficit without it, Finger Spin experiments extend the deficit beyond locomotion with heterogeneous severity across seeds, and shared-bank effect geometry is both candidate- and support-dependent. We also test practice-side questions. In DreamerV3 the posterior distribution, not its sample, carries the current query; ensemble disagreement ranks error only near training support; and a frozen support-aware score degrades held-out error ranking in both tested transfer directions while native disagreement remains informative in both. We conclude that intervention fidelity must be audited directly, capture-first, on the model's native interface.
Read the full abstract

Computer Vision

Discrete Diffusion Bridges for Spatiotemporally Aligned Image Translation and Generation

We propose Discrete Diffusion Bridges (DDB), a novel framework designed to resolve the fundamental spatiotemporal misalignment of standard discrete diffusion in image translation and generation. By corrupting data into a pure mask state via a random schedule, the conventional forward process induces a twofold misalignment: spatially, this pure-mask destination entirely discards the rich structural priors of the source image; temporally, the random masking order inherently contradicts the “easy-first, hard-last” decoding mechanism used during inference. To address this, DDB constructs a direct and efficient trajectory between domains. Spatially, we introduce a hybrid absorption mechanism that redefines the absorbing state to a stochastic mixture of mask and source tokens, effectively injecting source prior as spatial anchors into the latent space. Temporally, we design an information-guided noise schedule that quantifies semantic variation to prioritize the corruption of high-information regions at earlier timesteps. This ensures the model learns to resolve difficult semantic changes using robust context from invariant regions. Extensive experiments validate the versatility and robustness of our framework across diverse generative paradigms. DDB effectively balances edit alignment with structural fidelity across both text-guided semantic manipulation and pure structural image translation, while inherently complementing text-to-image generation and guaranteeing robust high-quality decoding under extremely low sampling steps. Code and models are available at \href{https://github.com/HKU-HealthAI/DDB}{https://github.com/HKU-HealthAI/DDB}.
Read the full abstract

All authors Xing Xie, Jiawei Liu, Shijun Zhou, Huijie Fan, Zhi Han, Yandong Tang, Liangqiong Qu

Computer Vision

Partition-Aware Unlearning for Removing Spurious Correlations in Large Vision-Language Models

Large Vision-Language Models (LVLMs) achieve strong performance across many multimodal tasks; however, they often exploit spurious object-background correlations, resulting in predictions driven by contextual shortcuts rather than object-relevant visual evidence. Despite growing interest in hallucination and robustness evaluation, existing benchmarks provide limited control over whether model predictions are grounded in the target object or induced by correlated background cues. In this work, we introduce PURGE (\underline{P}artition-aware \underline{U}nlearning for \underline{R}emoving spurious-correlation \underline{G}enerated \underline{E}rrors), a framework for constructing, benchmarking, and mitigating spurious-correlation-induced failures in LVLMs. The framework consists of: – (1) Structured dataset construction wherein we develop three complementary structured data construction strategies that partition examples by object-relevant evidence and spurious background cues, enabling controlled diagnosis of shortcut reliance; and – (2) Partition-aware unlearning, which uses these partitions to selectively remove spurious object-background associations while preserving object-based reasoning. We evaluate the \algo~framework across multiple LVLMs, including LLaVA-1.6-7B, Qwen3-VL-8B-Instruct, and Qwen3.5-9B, together with CLIP as a vision-language encoder, on a diverse suite of benchmarks, including CHAIR, POPE, Causal-HalBench, MM-SpuBench, AMBER, MMHal, and Waterbirds. Our results show that PURGE consistently reduces hallucinations and spurious-correlation-driven errors while maintaining or improving overall performance in most evaluated settings, providing both a reusable evaluation protocol and an effective mitigation framework for more reliable LVLMs.
Read the full abstract

All authors Aditi Sarker, Nazreen Shah, Rafi Ibn Sultan, Rhongho Jang, Dongxiao Zhu, Prashant Khanduri

Computation and Language

Generating Clinical Vignettes that Preserve Cognitive Formulations

Large language models can generate fluent clinical case vignettes, but fluency alone does not ensure fidelity to a specifiable clinical structure. We introduce FORMA, a theory-grounded framework that compiles a cognitive model of a disorder into a directed weighted graph, samples a person-specific configuration of that graph, and validates whether the generated vignette preserves the specified components and causal links. We instantiate FORMA on Posttraumatic Stress Disorder using the Ehlers and Clark cognitive model, generating 16,500 vignettes across 500 personas, 11 generation models, and three ablation conditions. Evaluation combines an external edge-recovery probe, two clinical experts, a scaled LLM judge, and a clinician user study with 100 licensed practitioners. The cognitive graph is recoverable from full-condition vignettes (MCC = +0.41, AUC = 0.70) but not from zero-shot generation (MCC = +0.01, AUC = 0.50). Experts rate full vignettes substantially higher than zero-shot alternatives, and clinicians perceive them to be human-written 85% of the time, compared with 22% for zero-shot. FORMA also reduces demographic disparity in perceived quality by 1.5-7x. These results show that cognitive formulation can serve as an auditable specification for scalable synthetic clinical text generation. A repository with the data and code is available online: https://github.com/Amit-Oren/FORMA.
Read the full abstract

All authors Amit Oren, Nimrod Hertz-Palmor, Dean Ariel, Guy Laban

Materials Science

Van Hove singularities at the $L$-face of the lutetium nitride phonon dispersion

We report the structural and vibrational properties of the prototypical 4$f$-filled nonmagnetic member LuN of the lanthanide nitrides, LnN, with elastic and inelastic neutron scattering data at $4$~K. We find a peak in the generalized density of states which, through input from a DFT+$U$ computation, we ascribe to a van Hove singularity on the fourfold-degenerate $L$-face of the Brillouin zone. This work advances the understanding of phonon dynamics in LnN beyond the $\Gamma$-point.
Read the full abstract

All authors M. Markwitz, K. Kneisel, R. G. Buckley, K. C. Rule, M. Chegeni, H. J. Trodahl, W. F. Holmes-Hewett, J. D. Miller, F. Natali, M. Maddah, B. J. Ruck, S. Granville

Databases

Demystifying and Improving Lazy Promotion in Cache Eviction

Cache eviction algorithms play a critical role in the performance of modern data systems, yet their scalability is often limited by the high computational overhead associated with object promotions. Lazy Promotion techniques have emerged as relaxations of traditional Least-Recently-Used (LRU) methods, designed to alleviate lock contention and increase throughput. This work uses production traces from real-world systems to benchmark five Lazy Promotion strategies: Probabilistic-LRU, Batch-LRU, Delay-LRU, FIFO-reinsertion, and Random-LRU. We evaluate these techniques across miss ratio, scalability, promotion count, and a novel metric called promotion efficiency, which measures the number of hits per promotion. Our results reveal that Delay-LRU and FIFO-reinsertion significantly improve promotion efficiency, whereas Batch-LRU and Probabilistic-LRU struggle to reduce promotions without significantly increasing miss ratio. We further explore the impact of lazy promotion in advanced algorithms such as ARC and 2Q and make a similar observation. Moreover, we uncover substantial optimization potential, showing that most cache promotions are unnecessary when equipped with oracle knowledge. To further reduce promotions in LRU, we propose two novel enhancements-Delayed FIFO-reinsertion (D-FR) and Age-Guided Eviction (AGE)-that reduce promotions by 20-60% while achieving a similar or lower miss ratio.
Read the full abstract

All authors Qinghan Chen, Muhammad Haekal Muhyidin Al-Araby, Ziyue Qiu, Zhuofan Chen, Rashmi Vinayak, Juncheng Yang

Computer Vision

SVI2LoD3: Agent-Driven Reconstruction of LoD3 Facade Openings in Semantic 3D City Models from Volunteered Street View Imagery using Large Language and Visual Models

This paper presents an end-to-end, agent-driven pipeline for the LoD3 reconstruction of facade openings in 3D city models, producing directly usable CityGML-conform outputs. In contrast to existing approaches that rely on supervised semantic segmentation and therefore require large amounts of manually annotated training data, the proposed method employs a zero-shot segmentation strategy. This substantially reduces the annotation effort while still achieving strong performance in our benchmark on the eTRIMS dataset. A further key contribution is the enforcement of correct partonomic hierarchies, thereby producing CityGML-conform LoD3 building models. Beyond the reconstruction pipeline itself, this work also introduces a novel evaluation metric for facade reconstruction, termed Facade Feature Distance (FFD). Unlike conventional metrics such as mIoU or FRDS, which assess similarity primarily through pixel-wise overlap, FFD measures distance in a high-level feature space derived from a vision transformer. In doing so, it captures both semantic correctness and architectural layout, providing a more suitable assessment of facade reconstruction quality. The proposed pipeline and evaluation strategy together offer a practical and scalable contribution toward the automated generation and analysis of semantically enriched 3D city models. The developed code is published at: https://github.com/hcu-cml/citydb-SVI2LoD3-ai.
Read the full abstract

All authors Elmehdi Kanna, Lukas Arzoumanidis, Huynh Duc An Son Nguyen, Youness Dehbi

Statistical Mechanics

Asymmetry-controlled resonant transport in a Brownian flashing ratchet

We investigate directed transport in a one-dimensional Brownian flashing ratchet with a piecewise-linear asymmetric periodic potential. Numerical solutions of the Fokker–Planck equation and Brownian dynamics simulations reveal a nonmonotonic dependence of the stationary current on the switching frequency, with a resonant maximum whose position depends on the potential asymmetry $\delta$ and barrier height $\Delta$. For moderate asymmetry, $|\delta|<0.5$, the current obeys a scaling form that separates the dependence on the potential parameters from a common frequency dependence, resulting in a data collapse upon appropriate scaling of the current and frequency. The current amplitude varies linearly with $\delta$, while the resonance frequency follows $\nu(\delta,\Delta)=\nu_0(\Delta)/ [1-b_0\, \delta^2]$. We interpret the resulting $(1-\delta^2)^{-1}$ scaling in terms of coupled relaxation along the two branches of the asymmetric potential, which provides a physical basis for the observed dependence of the resonant frequency on the potential asymmetry.
Read the full abstract

Computation and Language

Beyond Fluency: A Rubric-Based Benchmark for Evaluating Saudi Dialect and Cultural Competence in Large Language Models

Large language models are increasingly deployed in Arabic-speaking markets, yet standard benchmarks overwhelmingly reward Modern Standard Arabic (MSA) fluency while leaving dialectal and culturally grounded competence unmeasured. This gap is consequential: everyday Arabic is largely dialectal, and dialect encodes social meaning that MSA-centric evaluation cannot capture. We present a rubric-based benchmark for the Saudi dialect, comprising 31 expert-authored prompts spanning idiomatic, pragmatic, lexical, and culturally-embedded phenomena, each paired with an expert-established ground truth. Our methodology separates evaluation into a model-agnostic phase, in which atomic, MECE positive criteria are derived solely from the ground truth, and a model-specific phase, in which four state-of-the-art systems – Claude Opus 5, Gemini 3.7, GPT-5.6, and Kimi K3 – are scored against those criteria and penalised for errors they actively introduce. Across 124 model-prompt evaluations we catalogue 466 error instances under a nine-category taxonomy. The four systems cluster within a narrow macro-average band (42.7%-53.1%), with no model exceeding 55% and every model recording at least one negative-scoring prompt, confirming that Saudi dialectal competence remains broadly unsolved. Notably, Ambiguous Framing is the dominant failure mode (37.3% of errors) while outright Hallucination accounts for only 11.2%, indicating that models fail less by stating falsehoods than by distorting register and flattening pragmatic nuance. We further observe a consistency-versus-ceiling trade-off and model-distinctive error signatures. We release the full prompt set, ground truths, and scored rubrics to support reproducible dialectal evaluation.
Read the full abstract

All authors Ghassan Al-Sumaidaee, Sajjad Abdoli, Ahmed Rashad, Maxim Legg

Symplectic Geometry

Taut foliations through a contact lens

This survey explores the rich interplay between (taut) foliations and (tight) contact structures in dimension three, highlighting recent work of the author. We outline a new method for constructing taut foliations from suitable pairs of contact structures, and discuss some applications and future research directions. In particular, we give a brief account of work in progress with Jonathan Zung on ziggurats for taut foliations transverse to pseudo-Anosov flows, and we propose a (mostly speculative) contact perspective on the $L$-space conjecture.
Read the full abstract

Artificial Intelligence

AutoCRAT: Within-trajectory Joint Control of Stochasticity and Compute for LLM Reasoning

Large language models (LLMs) achieve strong reasoning performance, which depends critically on inference-time decisions. Yet these decisions are commonly handled by static, one-size-fits-all policies, limiting adaptation to diverse tasks and reasoning stages. Recent adaptive methods partially address this limitation, but they primarily adapt either decoding stochasticity (how the model explores) or reasoning compute (how long the model reasons) in isolation, leaving their interaction within a single reasoning trajectory unmodeled. To address this challenge, we shift toward a within-trajectory joint control view, and instantiate it in AutoCRAT, a decoder-side controller for frozen backbones. Using only signals available during decoding, AutoCRAT jointly adjusts sampling stochasticity and reasoning budget during generation. AutoCRAT operates over a discrete action space and updates control decisions only at semantic boundaries, improving stability while remaining responsive to the evolving reasoning process. Comprehensive evaluation across 6 benchmarks demonstrates that AutoCRAT (I) uses 13.8-52.7% fewer inference tokens on average than recommended static configurations, (II) surpasses recommended static and adaptive baselines by 1.5-4.5% in relative accuracy, and (III) enjoys strong cross-backbone transferability.
Read the full abstract

All authors Hanjun Luo, Qiushi Liu, Jingya Zhang, Haihong Pang, Jiaheng Wen, Yifei Ma, Yu Yao, Chengxi Zhang, Hanrong Zhang, Yankai Chen, Hanan Salam

Sound

How Well Do Generative Music Models Follow Emotion Conditioning?

Recent generative music models offer increasingly fine-grained control through text and audio conditioning, yet how faithfully they follow intended emotional cues remains an open question. We address this gap with a unified evaluation pipeline for emotion-following in generated music. Using all 1000 tracks in GTZAN, we extract semantic audio descriptions with DashengLM, an audio captioning model, and estimate source-track valence and arousal with Music2Emotion, a music emotion recognition model. We construct affect-aware text prompts by combining descriptions with top-ranked emotion tags and generate 30-second outputs with three systems, Stable Audio Open, MusicGen, and InspireMusic, evaluating both text- and audio-conditioned generation. To measure emotion-following, we compute valence and arousal on generated audio and compare them with the source tracks using absolute error and Euclidean distance in valence-arousal space. Text-conditioned generation consistently outperforms audio conditioning, with MusicGen (text) and InspireMusic (text) achieving the best performance, while audio-conditioned variants prove less stable. We further find that valence is preserved more reliably than arousal and that emotion-following varies substantially across genres. These findings underscore the importance of evaluating affective controllability directly rather than relying solely on general quality or prompt-relevance metrics.
Read the full abstract

Geometric Topology

Khovanov-Rozansky homology over $\mathbb{F}[U,V]$

We define a 'full' version of Khovanov-Rozansky homology: a chain complex CH(K) over F[U,V] whose chain homotopy type is a knot invariant and which has the property that setting U = V = 0 recovers the reduced Khovanov-Rozansky homology. We explore the algebraic structure of CH(K) and show that a variation of the structure theorem for knot Floer homology applies. This allows us to define counterparts to knot Floer concordance invariants $\tau$, $\epsilon$ and $\phi_j$ in the Khovanov-Rozansky setting.
Read the full abstract

Numerical Analysis

A variant of the block preconditioner for indefinite complex symmetric linear systems

In this paper, we propose an efficient preconditioner for solving indefinite complex symmetric linear systems within a block preconditioning framework. We analyze the convergence of the corresponding iterative method and investigate several spectral properties of the preconditioned matrix, including eigenvalue distributions and eigenvector structures. The new preconditioner is used to accelerate the convergence of the flexible version of GMRES. Numerical experiments are presented to illustrate the effectiveness of the proposed preconditioner, and comparisons with existing block preconditioners demonstrate its superior performance.
Read the full abstract

Analysis of PDEs

On Cauchy Problems for Parabolic Equations with Rough Coefficients

The presented work investigates the Cauchy problems for parabolic equations in both non-divergence and divergence forms with rough diffusion coefficients, which commonly arise in composite media, financial pricing, and viscous fluids. Under critical regularity settings, we establish the unique solvability in optimal fractional Sobolev spaces. The first key technical step lies in constructing an effective approximation scheme with truncated and mollified diffusion coefficients $a^\epsilon(t,x)$, for which we rigorously prove the uniform preservation of high-frequency smallness. By incorporating this approximation scheme with paraproduct decomposition, Fefferman-Stein maximal inequalities, Coifman-Meyer bilinear estimates, and refined Sobolev embeddings, we close the uniform a priori estimates and then pass to the limit. Furthermore, we prove that the threshold $s < \frac{1}{2}$ is sharp by constructing explicit counterexamples. These theoretical results provide a rigorous mathematical framework in the study of Cauchy problems for parabolic equations with rough coefficients.
Read the full abstract

Machine Learning

Robust Broad Learning System with Wave Loss for Classification under Data Uncertainty

Broad Learning System (BLS) offers an efficient alternative to deep architectures by enabling fast learning through randomized feature mapping and closed-form solutions. However, its reliance on squared error loss makes it highly sensitive to noise, outliers, and corrupted labels, limiting its reliability in real-world scenarios. To address this limitation, we propose Wave-BLS, a robust broad learning framework that integrates the wave loss function, which is asymmetric, bounded, and smooth, enabling controlled penalization of large errors. The proposed formulation replaces the standard least-squares objective with a wave-loss-based optimization problem, solved efficiently using a Nesterov accelerated gradient (NAG)-based scheme without requiring matrix inversion, thereby improving scalability. Extensive experiments on 30 UCI benchmark datasets demonstrate that Wave-BLS consistently outperforms classical BLS and several robust variants. Statistical validation using Friedman and Nemenyi post-hoc tests confirms the significance of the observed improvements. Furthermore, robustness evaluations under controlled noise and outlier injection reveal that Wave-BLS exhibits substantially slower performance degradation compared to BLS, even in challenging contamination settings. These results establish Wave-BLS as a stable and robust alternative to existing broad learning models for learning under data uncertainty.
Read the full abstract

All authors Mushir Akhtar, A. Varshney, A. Quadir, A. Rahaman, M. Tanveer, Mohd. Arshad

Combinatorics

Equivariant Analytic Spectral Invariants of Aç{\i}kmeşe Lifts of Graphs with Self-Loops

We introduce and study equivariant analytic spectral invariants associated with the Aç{\i}kmeşe lift of a graph with self-loops $G_S$. The canonical $\mathbb{Z}_2$-action yields an orthogonal isotypic decomposition of the lifted Laplacian into the anti-symmetric part $\mathcal{L}(G_S)$ and the symmetric part $M_{sym}$, from which we prove that the spectrum of $\mathcal{L}(G_S)$ is exactly the even-indexed spectrum of the lifted Laplacian. We express twisted moments of these blocks as traces of the generalised twisted moment operator against the corresponding $\mathbb{Z}_2$-projections, and obtain a tight upper bound under certain restriction with a characterization of the equality case. We introduce the equivariant heat character and establish a trace-norm stability estimate. For the matrix element of the resolvent $(pI+\mathcal{L}(G_S))^{-1}$ associated with the loop vector, we derive a Laplace-transform identity for the equivariant heat character, a determinant formula, a spectral representation, and a Laurent expansion. We further characterize the case where $S$ is a union of connected components through resolvent and equivariant heat-character identities, and obtain explicit equivariant heat characters for joins of full-loop graphs and the line graph of a full-loop connected regular graph. Finally, we introduce the regularized equivariant heat integral and derive spectral and trace formulas for it.
Read the full abstract

Combinatorics

Recovering Laplacian Lattices from $L$-Functions of Graphs

We introduce $L$-functions associated with characters of the Jacobian of a finite graph, as a graph-theoretic analogue of the $L$-functions arising from unramified coverings of algebraic curves. These $L$-functions are defined using the Riemann–Roch structure on the graph and extend Lorenzini's two-variable zeta function. We show that if two connected graphs without bridges have isomorphic Jacobians and their $L$-functions agree under the induced correspondence of characters, then their Laplacian lattices coincide. We also show that these $L$-functions satisfy a functional equation. Finally, we give examples showing that neither the Jacobian nor the Lorenzini zeta function alone determines the Laplacian lattice.
Read the full abstract

Computer Vision

FIS-OT: Feature-Induced Optimal Transport for Unsupervised Action Segmentation

Unsupervised action segmentation is a challenging task. It involves finding action categories and boundaries in videos without labels. Existing Optimal Transport (OT) methods use global constraints. This causes them to overlook the use of local information. Furthermore, existing Optimal transport architectures are prone to confirmation bias because they overly trust the pseudo-labels they generate. This causes models to learn from noise in the early training stages. To address these issues, we propose FIS-OT. It is a novel Feature-Induced Structured Optimal Transport framework. First, we introduce a Feature Enhanced Generator (FEG) module. It serves as an internal regularizer. By using triplet loss, FEG captures local consistency. It provides robust supervision that is independent of noisy pseudo-labels. Second, we propose a Feature-Induced Residual Structural Prior. This combines a fixed temporal backbone with dynamic feature similarities. This design ensures temporal continuity. It also allows the solver to adapt to complex action structures. Finally, we establish a cyclic optimization loop. This aligns local feature learning with global structural alignment. Extensive experiments on the three datasets show the effectiveness of our method.
Read the full abstract

All authors Linxiang Peng, Xinyao Qin, Jinhan Li, Di Yang, Jiangtao Wang

Cryptography and Security

Breaking Ambient Trust: In-Network Per-Process Access Control Against Lateral Movement

Enterprise networks remain vulnerable to Advanced Persistent Threats (APTs), where adversaries gain an initial foothold and move laterally across the network, accumulating access permissions hop by hop to reach critical targets. Existing network defenses cannot track user movement at the process level across the network; instead, they grant ambient trust to all processes within a host. As a result, once a host is compromised, malicious processes inherit the victim's permissions, thereby expanding the attacker's access scope and enabling further lateral movement. To address this gap, we present NetZone, an in-network access control that confines each user process to a fixed access scope that persists as the user moves across the network. NetZone introduces a new abstraction, called AccessScope, which represents a lightweight access capability bound to the user's processes. Each AccessScope encodes the set of hosts a user identity is authorized to access and is embedded in the process's outgoing network traffic for validation before reaching its destination. As users pivot across hosts, AccessScope propagates with their traffic, rebinds to the receiving process, and persists across hosts. This ensures that regardless of network location, the user's processes are consistently governed by their bound AccessScope and their access permissions remain unchanged. To handle the high volume of network traffic generated by processes, we develop a data-plane co-design that integrates programmable switches with eBPF. NetZone employs a set of in-network optimizations and lightweight AccessScope persistence techniques to inspect the embedded AccessScope on the fly, enabling line-rate processing of high traffic volumes with negligible latency overhead. Our extensive evaluations show that NetZone can effectively defend against sophisticated attack scenarios without introducing noticeable overhead.
Read the full abstract

Computation and Language

Evolutionary Soups: Evolving Mixture-of-Experts for Multi-Objective LLM Alignment

Large language models are increasingly required to generate responses that satisfy multiple competing objectives. Since optimal trade-offs depend on both user preferences and input prompts, controllable multi-objective generation must dynamically adapt models at inference time without retraining. To address this, we propose Evolutionary Soups, a mixture-of-experts framework for fine-grained generation control, with gating networks trained via an evolutionary algorithm. The per-layer gating networks dynamically produce expert-merging coefficients from hidden-state representations, while the evolutionary algorithm incorporates greedy hypervolume contribution for effective evolution of these gating networks, achieving consistent improvements on large and noisy training datasets and broader coverage of the non-convex Pareto front. Experiments across three tasks demonstrate the effectiveness of Evolutionary Soups over baselines: it achieves the best hypervolume, linear utility, and Tchebyshev utility (~20% improvement) among controllable methods on all tasks.
Read the full abstract

All authors Lingxiao Kong, Steffen Staab, Cong Yang, Oya Beyan, Zeyd Boukhers

Cryptography and Security

A New Algebraic Algorithm for LWE

The Learning With Errors (LWE) problem, introduced by Regev in 2005, is central to modern cryptography and post-quantum security. The algorithms to solve the search version of the problem, Search-LWE, can be broadly categorised into algebraic, combinatorial and lattice-based. In this work we propose a new algebraic algorithm for the Search-LWE problem. At a high level, the algorithm combines linear-algebraic techniques with S-polynomial-based methods from Groebner basis computation. We provide a direct complexity analysis of our algorithm, avoiding semi-regularity assumptions and complexity bounds derived from the degree of regularity. Our algorithm achieves a polynomial improvement in complexity over prior results that use Groebner basis methods to solve Search-LWE.
Read the full abstract

Functional Analysis

The Pego Theorem for the Hilbert–Schmidt Class

This paper establishes an operator-theoretic version of Pego's compactness theorem within the framework of quantum harmonic analysis on general locally compact abelian phase spaces. We show that a bounded set of Hilbert-Schmidt operators is precompact if and only if it is uniformly equicontinuous under phase-space shifts and its Fourier-Weyl transform is uniformly equicontinuous on the dual phase space. We provide applications to quantum physics.
Read the full abstract

Metric Geometry

Sylvester's four point problem for ball-convex bodies

We investigate Sylvester's classical four-point problem for ball-convex bodies, where convex hulls are replaced by intersections of balls of a fixed radius $R$. We show that the Efron–Buchta identities persist in this framework, so that the distribution of the number of vertices of the $R$-ball-convex hull of $n$ uniform random points is determined by the moments of the volumes of the hulls of its subsamples, and vice versa. For three and four uniform random points in an arbitrary planar ball-convex body, we determine the Sylvester probabilities of the $R$-ball-convex hull. We also give an extension of Groemer's inequality for the expected area of the ball-convex hull of $n$ random points and show that it is minimised by the disc of the same area. In particular, this yields an isoperimetric inequality for the ball-convex analogue of the affine length of the boundary curve. For the unit disc, we can express the Sylvester probabilities using dilogarithmic functions of the radius $R$, and give explicit values at $R=1$. As $R\to\infty$, these recover the classical Sylvester probabilities. As another consequence, we determine the exact distribution function of the circumradius of three uniform random points in he disc on $[1,\infty)$.
Read the full abstract

Computer Vision

SpanCalib-VLM: Calibrated Hallucination Span Detection in Vision-Language Models

Detecting hallucinations in Large Vision-Language Models (LVLMs) requires both accurate span localization and well-calibrated confidence scores. Fine-tuned generative VLMs excel at identifying hallucinated text spans but suffer from overconfidence and high inference latency. Discriminative sequence taggers offer deterministic speed and superior calibration but exhibit conservative span recall. We present SpanCalib-VLM, a hybrid dual-system for the SHROOM-Visions Shared Task that combines a multimodal sequence tagger, consisting of XLM-RoBERTa-Large fused with a SigLIP vision encoder via cross-attention, with our fine-tuned generative VLM (Qwen3.5-4B-SHROOM-SFT). Through a Union-Calibrated Fusion strategy, candidate spans from the generative model are re-scored with calibrated probabilities from the sequence tagger. On the SHROOM-Visions English validation split, our ensemble achieves a Pearson calibration correlation of 0.41 and an overall IoU of 0.39 (with a clean-response IoU of 0.91 and overall detection accuracy of 70.7%). On the official blind test set, SpanCalib-VLM achieves a Pearson correlation of 0.32 and an IoU of 0.25 on English, substantially outperforming the shared task baseline (0.15 on both metrics) and demonstrating consistent gains across all four languages. We make our model weights and code publicly available.
Read the full abstract