Peer Revenue Equity in science

Papers

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

All of arXiv

Friday 28 August

Differential Geometry

Flows and Homotopies of Banach Lie algebroids

We establish flow and homotopy tools for Banach Lie algebroids that do not rely on compactly supported extensions. Under suitable hypotheses, we construct parameter-dependent local extensions of sections along algebroid paths, complete lifts and their fibrewise-linear evolutions, and a Bochner-integral variation-of-constants formula. We then prove that an admissible-path homotopy is equivalently a Lie algebroid morphism from the parameter square and a solution of the associated transport equation. This provides the extension, transport and infinitesimal variation theory required for a local integration program of Banach Lie algebroids to be developed in a sequel.
Read the full abstract

Software Engineering

Post-Edit Re-Verification in Simulator-Backed Engineering Agents: A Controlled Comparison of Verification-Cadence Guidance

Engineering agents that interact with external simulators may need to coordinate design modification with reacquisition of engineering evidence for the modified state. We ask whether first post-edit re-verification changes when explicit verification-cadence guidance is retained versus omitted while verification-relevant state/facts are held constant. Cadence-Guided (CG) retained an instruction to request a new simulation after a substantive modification, whereas Cadence-Omitted (CO) removed that instruction; neither condition used a hard gate. The study therefore measures instruction-conditioned post-edit verification-policy adherence rather than spontaneous recognition that prior evidence has become stale. Using DWSIM as the simulator backend and continuous valve-pressure adjustment, five Alibaba/Qwen models were evaluated on eight synthetic cases; each model-case-condition combination was executed three times via live API calls, yielding 120 evaluation slots per condition. Re-verification was observed in 94/120 CG slots versus 32/120 CO slots; cadence violations occurred in 26/120 versus 87/120; and bounded final success was reached in 95/120 versus 35/120. qwen3.5-35b-a3b showed minimal re-verification (1/24 in CG and 0/24 in CO) and no final success in either condition. Within this bounded protocol, explicit post-edit verification-cadence guidance was associated with more re-verification, fewer cadence violations, and more frequent bounded final success, supporting the treatment of verification cadence as an explicit interaction-protocol component.
Read the full abstract

Mesoscale and Nanoscale Physics

Topological signatures in the curvature-induced energy response

Relativistic effective field theory predicts a topological energy response to a gravitational field that appears at third order in spatial gradients. Here, we investigate how this response emerges in the nonrelativistic Haldane model using a microscopic lattice formulation of curvature-induced deformations. We find that the leading first-order energy response is nonuniversal and depends on the bond-resolved structure of the deformation; in particular, it vanishes for a symmetric modulation of the three nearest-neighbor hoppings. At third order, the curvature-induced energy response contains a bond-symmetric component together with nonuniversal bond-dependent contributions. The bond-symmetric component, which is isolated by a uniform modulation of the three hoppings, exhibits a universal discontinuity across the topological transition whose magnitude agrees with the relativistic gravitational Chern–Simons prediction. Thus, although the full curvature-induced energy response generally contains nonuniversal lattice contributions, its third-order response retains a universal topological component beyond the relativistic limit.
Read the full abstract

Computer Vision

Dual-Stream Semantic Guidance with Prototype Anchor Calibration for Source-Fully-Free Adaptation of Vision-Language Models

Source-Fully-Free Domain Adaptation (SFF-DA) has emerged as a strategic paradigm to adapt Vision-Language Models (VLMs) without any access to source data or task-specific source models. However, we identify a critical Dual Semantic Drift that hinders this process: static drift arising from the rigidity of fixed class embeddings, and dynamic drift stemming from the divergence of generated captions, causing severe semantic misalignment that intensifies the stability-plasticity dilemma. To address this, we propose DSSG (Dual-Stream Semantic Guidance), an end-to-end framework that reconciles fine-grained plasticity with global stability. Our core contribution is the Dual Semantic Guidance (DSG) module, which integrates a caption stream for domain-specific knowledge with a class-anchor stream to anchor global categorical consistency. Furthermore, a Dynamic Cross-Modal Knowledge Distillation (CMKD) module is introduced to leverage the evolving teacher distribution for calibrating teacher-student consistency. Building upon DSSG, we further introduce Prototype Anchor Calibration (PAC), yielding DSSG-PAC, which periodically calibrates prototype anchors and caches them until the next calibration. This design reduces redundant text-side computation while preserving the adaptability of class guidance to the evolving text space. We further establish SFF-DA risk bounds that relate student risk to semantic-teacher quality and teacher–student discrepancy. Extensive experiments demonstrate that DSSG consistently outperforms current state-of-the-art methods across multiple benchmarks, while DSSG-PAC largely preserves its adaptation performance with 18.9% lower total adaptation time. The code is available at https://github.com/mrmenand/DSSG.
Read the full abstract

All authors Weiwei Xiang, Shun Peng, Guangyi Xiao, Hao Chen, Lei Yang

Artificial Intelligence

The Shape of Power: A Multilingual Framework for Social Power Reasoning in Dialogues

Social power plays a fundamental role in shaping human interaction, yet computational studies of power remain limited to narrow linguistic and cultural settings. Existing datasets further lack the demographic and relational depth needed for robust cross-cultural analysis. To address this gap, we introduce a theoretically grounded framework for studying social power in naturalistic multilingual dialogue through movie screenplays. The framework integrates a schema informed by social science theory, a native speaker annotation pipeline refined through pilot studies, and a custom interface for scalable cross-lingual analysis. Using this framework, we constructed an initial corpus containing 15,836 annotated instances from 100 scenes in French and Egyptian Arabic movies. Our analysis reveals strong agreement on observable demographic and contextual attributes, while socially interpretive aspects, such as power asymmetry and intention alignment, remain more contested, highlighting the complexity of social power across cultures. We evaluated 6 Large Language Models (LLMs) and Multimodal LLMs on cross-cultural social power reasoning, finding persistent gaps between human and model agreement in relational and theory-of-mind reasoning. Our work introduces the first extensible multilingual framework for studying social power in dialogues and provides an initial evaluation setting for studying cross-cultural social reasoning.
Read the full abstract

Materials Science

Light-induced atomic motion in ionic crystals

Atomic motion in solids is conventionally driven by elastic collisions between ionizing particles and atoms, which transfer momentum and induce lattice displacements. In this work, we demonstrate a different mechanism for atomic displacement based on optical excitation of scintillating ionic crystals. Ionic crystals are unique systems because of the closed-shell electronic configuration of their constituent ions. In these materials, excitation above the band gap generates a hole that strongly distorts the lattice, resulting in the formation of a self-trapped hole (STH). The STH is Coulomb-attracted to the electron, thereby forming a self-trapped exciton (STE). Here, we demonstrate that in BaF2 - one of the fastest scintillators - the STE structure promotes the formation of long-lived electron and hole traps that persist in the lattice at room temperature. Such trapped electron-hole pairs occupy vacancy-interstitial fluorine pair positions, and can be created indiscernibly using optical or ionizing radiation excitation, as long as the STH is formed. Further, we demonstrate that it is possible to control the defect evolution with light. Selective optical stimulation of the trapped electrons or holes enables the regeneration of the STE at later times. This light-controlled defect engineering allows us to increase the yield of the STE signal appearing as optically stimulated luminescence (OSL) and to image the spatial distribution of the initial energy deposition, holding strong potential for ionizing-radiation detection. These findings provide a common framework underlying scintillation and OSL in ionic crystals of the fluorite structure, allowing for optical manipulation of atomic vacancies-interstitial pairs in similar systems.
Read the full abstract

All authors Jacob C. Warming, Simon P. S. Jessen, Emma A. Husted, Jan Thøgersen, Stefan Gundacker, Brian Julsgaard, Peter Balling, Rosana M. Turtos

Machine Learning

Conditional Diffusion Models for Energy-Efficient Driving

Electrification of commercial delivery fleets is shifting fleet routing from distance- and time-based optimization toward energy-aware decision-making. Existing sequence models primarily provide deterministic point estimates or limited uncertainty summaries, which do not capture the range of plausible energy-consumption trajectories required for operational decision-making. In this work, we introduce a conditional diffusion framework that generates EV battery-current profiles conditioned on route features such as vehicle velocity and ambient temperature. The model combines a latent conditioning encoder with a temporal 1D U-Net denoising backbone that enables trip-related conditions to be mapped into a shared representation and guides the reverse diffusion process. We evaluate the framework on an open-access commercial EV telemetry dataset containing 12k trips from 9 vehicles. The proposed latent-conditioned diffusion model generates realistic cur- rent trajectories that capture both the dominant temporal envelope and sharp transient events. The model achieves a Wasserstein distance of 0.0029 between generated and measured current distributions below the real vs real reference distance of 0.0085 indicating that generated samples lie within the empirical variability of the test set. We further demonstrate that learned latent conditioning substantially improves performance over direct condition injection, reducing the Wasserstein distance by 89.1% and MAE by 52.8%. This work demonstrates a generative modeling framework for characterizing EV energy consumption under real-world operating conditions, providing an essential foundation for uncertainty-aware fleet planning in large-scale operational settings.
Read the full abstract

All authors Hemanth Neelgund Ramesh, André Snoeck, Chyi-Fu Hong, Shijing Sun

Mathematical Physics

Lattice Green's function of the hyperkagome lattice: modular uniformization at level 30 from an orthogonal differential Galois group

The lattice Green's functions of the cubic lattices are classical: each is a symmetric square of a second-order operator, with a closed form in complete elliptic integrals. The hyperkagome lattice, realized by the iridium sublattice of the spin-liquid candidate Na4Ir3O8, has resisted such a treatment: Varma and Monien reduced its density of states to a threefold integral they found no way to solve exactly. From exact lattice moments we obtain an irreducible third-order linear differential (Picard-Fuchs) operator annihilating the Green's function. It is not a literal symmetric square, yet its symmetric square carries a rational solution: the differential Galois group is orthogonal, and an order-two differential intertwiner makes the operator projectively a symmetric square. Our main result is that the underlying second-order operator is the uniformizing equation of the genus-zero modular curve of Gamma_0(30)+, the level-thirty group generated by Gamma_0(30) and all its Atkin-Lehner involutions, an explicit eta quotient parametrizing the natural variable. That uniformizing equation is itself not new - it is a row of the Lian-Yau table of genus-zero groups - so what is new here is the identification of a lattice Green's function with it, together with a proof: the Schwarzian identity is established by an a priori pole-degree bound rather than checked numerically, which also proves the tabulated entry. The hyperkagome Green's function is therefore modular at level thirty, supplying the closed form Varma and Monien sought. Its weight-two period is obtained explicitly: a depth-one quasimodular form which we prove is not a modular form times an algebraic function at any weight.
Read the full abstract

Robotics

Contact-Guided Exploration for Non-Prehensile Locomanipulation with Multi-Critic RL

Non-prehensile manipulation offers versatile skills for moving and rearranging heavy or bulky objects, particularly when combined with a mobile manipulation platform. However, both model-based and model-free approaches struggle with the complex hybrid dynamics and the sparsity of the contact in these tasks. To address these challenges, we propose a contact-guided exploration strategy implemented within a Multi-Critic Reinforcement Learning (RL) framework. A dedicated exploration critic is trained with a dense contact-seeking reward that guides the end-effector toward meaningful contact points; its influence is progressively decayed to recover a task-optimal policy. We obtain candidate interaction points from a general-purpose grasping algorithm, enabling the exploration mechanism to generalise across various object geometries. We evaluate the approach on multiple tasks, including box pushing, chair transportation, and a dishwasher opening task. Finally, we validate the chair transportation policy through extensive experiments on a quadrupedal mobile manipulator, demonstrating deployable non-prehensile manipulation in the real world.
Read the full abstract

All authors Simone Tolomei, Mayank Mittal, Franco Angelini, Manolo Garabini, Paolo Salaris, Marco Hutter

Physics and Society

China's Shrinking Home Bias and Rising Disruptive Impact: Evidence from a Global Citation Network Analysis

China has become the world's largest producer of scientific publications, yet concerns persist that this growth is inflated by excessive domestic citation practices. In this study, we analyze a citation network of over 45 million publications from Web of Science (1980-2025) to investigate China's home citation bias and research impact. Using a network reshuffling null model to control for the structural effect of publication volume, we find that China's home citation bias is less pronounced than commonly assumed and has been steadily declining over the past two decades. Chinese researchers do not exhibit a significantly stronger home citation preference than other major countries, indicating increasing internationalization rather than insularity. Furthermore, using the persistent disruption framework, we show that Chinese papers are converging toward American papers in their capacity to produce paradigm-shifting work. These findings challenge prevailing narratives about Chinese scientific home bias and suggest that China's advances in research impact.
Read the full abstract

Computer Vision

Token-Budget Distillation: Transferring Full-Token Semantics to Compressed Video Vision-Language Models

Adapting video vision-language models (VLMs) is computationally expensive because video inputs produce a large number of visual tokens, making both fine-tuning and inference costly. Although visual token compression can reduce this overhead, direct adaptation on compressed inputs often causes semantic drift and noticeable performance degradation. We present Token-Budget Distillation (TBD), a parameter-efficient fine-tuning framework for adapting video VLMs under a fixed token budget. TBD freezes the pretrained backbone, updates only LoRA adapters, and integrates FlashVID-based visual token compression into the video pathway. To preserve full-token semantics under compression, TBD employs a dual-path teacher-student design, where a full-token teacher provides stable supervision and a compressed student is optimized with task loss, answer-region KL distillation, GT-anchored margin distillation, and reliability-aware KD control. This design enables the student to recover the semantic behavior of the full-token model while remaining efficient under aggressive token reduction. We evaluate TBD on three video VLM backbones, including LLaVA-Video, LLaVA-OneVision, and Qwen3-VL-8B-Instruct, across four video understanding benchmarks. TBD consistently outperforms compression-only baselines under both moderate and aggressive compression. On LLaVA-Video at retention ratio R = 10 percent, TBD preserves 97.0 percent of the Vanilla model's average accuracy; on LLaVA-OneVision at R = 10 percent, it achieves an average score of 58.4 and matches 100.0 percent relative accuracy.
Read the full abstract

All authors Xiaoyang Guo, Guoping Luo, Jusheng Zhang, Keze Wang, Wenhao Wang

Image and Video Processing

CheXtriev: Anatomy-Centered Representation for Case-Based Retrieval of Chest Radiographs

We present CheXtriev, a graph-based, anatomy-aware framework for chest radiograph retrieval. Unlike prior methods focussed on global features, our method leverages graph transformers to extract informative features from specific anatomical regions. Furthermore, it captures spatial context and the interplay between anatomical location and findings. This contextualization, grounded in evidence-based anatomy, results in a richer anatomy-aware representation and leads to more accurate, effective and efficient retrieval, particularly for less prevalent findings. CheXtriv outperforms state-of-the-art global and local approaches by 18% to 26% in retrieval accuracy and 11% to 23% in ranking quality. The code is available at https://github.com/cvit-mip/chextriev.
Read the full abstract

Solar and Stellar Astrophysics

On the theoretical instability strips of {\gamma}-Doradus stars including the effect of the metallicity and rotation

Recent space missions such as Kepler have provided large-scale observations of the {\gamma}-Doradus instability strips (IS), which can be used to constrain models and explore their limitations. One persistent limitation is the prediction of the blue edge of the {\gamma}- Doradus IS, where a significant number of {\gamma}-Doradus stars are observed. Despite these observational advances, no systematic study has been undertaken to explore the effects of different physical processes on the {\gamma}-Doradus. Our aim is to systematically explore the theoretical {\gamma}-Doradus IS with modern tools, accounting for the effects of rotation and metallicity, providing a large grid of models and their oscillation parameters for the scientific community. We investigated the non-adiabatic pulsation properties of stars in a grid of stellar models with masses between 1.35 M{_\odot} and 2.5 M{_\odot}, metallicities between Z= 0.01 and Z= 0.025, and solid-body rotation rates ranging from {\Omega} = 0 to 0.5 {\Omega_{crit}}, where {\Omega_{crit}} denotes the critical rotation rate. We find that, across all computations, the theoretical {\gamma}-Doradus IS agrees well with the observed IS, except in the extended blue region. In terms of radial orders, our models consistently reproduce the excited modes in broad agreement with observations for l= 1, l= 2, and Rossby modes. We also show that the range of excited radial orders is strongly dependent on the effective temperature. The range of excited radial orders can be used to constrain the position of a star within the {\gamma}-Doradus IS and may provide insights into the physical mechanisms responsible for the discrepancies observed in the blue region. All computed grids are made available and include mode damping and growth rates for {\gamma}-Doradus stars, obtained with a time-dependent treatment of the convection-oscillation interaction.
Read the full abstract

Numerical Analysis

Stochastic trace estimation for positive trace-class operators

Implicit trace estimation aims to approximate the trace of a matrix or linear operator accessible only through matrix-vector or operator-vector products. In the matrix setting, the Girard-Hutchinson estimator typically requires $\mathcal{O}(\varepsilon^{-2})$ products to achieve accuracy $\varepsilon$, while the variance-reduced Hutch++ algorithm reduces this sample complexity to $\mathcal{O}(\varepsilon^{-1})$ for positive semidefinite matrices. We develop infinite-dimensional analogues of these estimators for positive trace-class operators on separable Hilbert spaces. The idealized estimators use Gaussian random elements whose covariance is determined by the target operator, leading to unbiased operator versions of Girard-Hutchinson and Hutch++. We prove high-probability error bounds analogous to the finite-dimensional matrix results; in particular, idealized infHutch++ achieves $\mathcal{O}(\varepsilon^{-1})$ sample complexity. For practical computation, we introduce truncated implementations that restrict the random samples to finite-dimensional subspaces; for fixed sample budget, we show that truncated infHutch++ converges in distribution to its idealized counterpart as the truncation dimension tends to infinity. Numerical experiments with integral operators, density-of-states approximations, and spectral filtering for a radial Dirac operator show that these truncated estimators can achieve accuracy comparable to the ContHutch++ algorithm by Zvonek, Horning & Townsend while using lower-degree function representations and smaller internal discretizations in chebfun.
Read the full abstract

All authors Zvonimir Bujanović, Luka Grubišić, Daniel Kressner, Hrvoje Olić

Machine Learning

Learning to Difference: Adaptive Reversible Differencing (AdaRDiff) for Time Series Forecasting

Reliable long-horizon time series forecasting is an important yet difficult problem. Trends and seasonality introduce complex temporal structure that challenges learning-based forecasting models. Differencing, which subtracts nearby past values to remove such structure, is the classical remedy, but its reliance on hand-picked orders and periods has kept it largely absent from recent deep architectures. We propose \textbf{\underline{Ada}}ptive \textbf{\underline{R}}eversible \textbf{\underline{Diff}}erencing (AdaRDiff), a generalized differencing approach that uses learnable weights to simplify the series through weighted differencing with previous time instants. This yields stabilized residuals on which forecasting is performed, after which the removed components are restored autoregressively to reconstruct the forecast, capturing trend and seasonality jointly through a single operator. This reconstruction admits a closed-form convolutional expression, which parallelizes on GPU and yields up to $33.7\times$ speedup over the naive recurrence. We furthermore rely on a two-phase training schedule that separates temporal structure discovery from reconstruction learning, as suggested by a theoretical analysis of the gradient when using a linear forecasting model. AdaRDiff attains state-of-the-art forecast accuracy across eight benchmarks spanning electricity, weather, traffic, and energy, at negligible parameter cost. Furthermore, it is designed as a plug-and-play module: integrating AdaRDiff improves eight diverse backbones, from linear models to Transformers, in the large majority of cases, by up to $25.9\%$ with a linear backbone and $18.3\%$ with iTransformer.
Read the full abstract

All authors Morad Laglil, Younes Hlal, Marouane El Hadari, Emilie Devijver, Eric Gaussier

Information Theory

Fine Difference Structure and Prime-Power Depth of Bent Partitions

A $p$-ary bent partition of $\mathbb{F}_p^n$ is a partition into $K$ nonempty cells such that every balanced assignment of its cells to $\mathbb{F}_p$ produces a bent function. It was asked whether every possible depth $K$ is a power of $p$; for general $p$, previous affirmative results required regularity or cell-symmetry hypotheses. We prove the stronger unconditional statement that, for every nonzero $h$, exactly $p^n/K$ points remain in the same fine cell under translation by $h$. Thus the fine cells form a partitioned difference family and the fine label map is zero-difference balanced. Consequently $K\mid p^n$, so $K=p^t$; nonempty cells further give $1\le t<n$. In even dimension, the classical cell-size theorem yields $K\mid p^{n/2}$. Together with the known odd-dimensional ternary three-fibre parameter restriction, this gives the global bound $t\le\lfloor n/2\rfloor$. The proof is an exact finite average over balanced coarsenings. The main counting identity and selected consequences are formalized and kernel-checked in Lean 4.
Read the full abstract

Analysis of PDEs

Observability for wave equations with critically singular potentials

In this article, we prove boundary observability estimates for wave equations on bounded domains $\Omega \subseteq \mathbb{R}^n$, with a critically singular potential that diverges as the inverse square distance to $\partial \Omega$. The result is applicable in all dimensions and for wave equations with general time-dependent lower order terms. The key geometric assumption is a convexity condition on $\Omega$. The main tool for our result is a global Carleman estimate that is carefully adapted to the critical singular nature of the potential.
Read the full abstract

Mathematical Physics

Reducing the domain of discrete multi-time determinantal point processes

We consider a class of discrete multi-time determinantal point processes whose correlation kernels admit a double-contour integral form, containing Schur processes as simple examples. We show that the form of the correlation kernel is preserved under conditioning of the point process to a restricted domain, however, with an integrand which becomes more complicated and which is characterized by a Riemann-Hilbert problem. We apply our general result to domino tilings of reduced Aztec diamonds and to lozenge tilings of hexagons with holes. Our results show a striking analogy with the Its-Izergin-Korepin-Slavnov method for integrable kernels.
Read the full abstract

Applied Physics

Beating the nonreciprocal isolation limit of integrated circulators by Floquet leakage interference

Time-modulation using semiconductor switches is a promising route to magnetless integrated nonreciprocal devices, as they can offer large modulation depths and high speeds. Yet, chip-scale devices are capped by the finite off-state capacitance of the switches, which induces detrimental leakage of the input wave to the isolated port, imposing a ceiling on the nonreciprocal isolation. Such leakage has largely constrained the development of nonreciprocal integrated systems at high frequency. Here, we beat this limit by using Floquet interference between leakages. In a time-Floquet switched-resonator circulator, two coherent leakages reach the isolated port: the release of the stored wave at the resonator's ring-down frequency, and the direct input leakage at the carrier frequency. We show that it is possible to create conditions under which the two leakages destructively interfere, and even completely cancel, yielding perfect isolation despite operating with non-ideal semiconductor switches. We experimentally confirm leakage interference in a 65-nm CMOS microwave Floquet circulator, reaching 40-dB isolation, which is more than 20 dB higher than the natural switch isolation. We also demonstrate that leakage interference has inherently fast dynamics, establishing itself within a single modulation period, allowing us to reverse the circulation chirality in 0.6~ns. Our results pave the way toward high-frequency integrated chips with ultra-high nonreciprocal isolation.
Read the full abstract

All authors Zhe Zhang, Haoye Qin, Junda Wang, Qiaolu Chen, Zhechen Zhang, Alireza Mafi, Ahsan Altaf, Romain Fleury

Optics

Performance Optimization of Short Reach Optical Interconnects based on Direct Detection

Short-reach optical interconnects are evolving toward data rates beyond 400 Gb/s per lane, driven by the bandwidth and energy-efficiency requirements of AI-enabled datacenter networks. At these operating speeds, channel impairments, device nonlinearities, and hardware constraints limit the effectiveness of conventional transceiver design and digital signal processing (DSP). This paper presents a unified framework for the optimization of direct-detection optical interconnects, encompassing digital surrogate modeling, receiver-side DSP optimization, and end-to-end (E2E) transceiver learning. The proposed formulation provides a common perspective for model-based and machine-learning-based approaches, including linear and nonlinear equalization, lookup tables, decision trees, neural-network receivers, and E2E optimization. Their performance is discussed together with computational complexity and hardware implementation aspects, highlighting the associated trade-offs. We show, through simulations and experimental validations, that for a 40~Gb/s 10~km link, decision trees can outperform by 0.5–1~dB conventional linear equalization, with negligible hardware requirements. Moreover, we show that an E2E technique based on transformers can provide a gain up to 6~dB, highlighting the potential of joint transceiver optimization to improve the performance of next-generation short-reach optical links.
Read the full abstract

All authors Luca Poì, Stella Civelli, Marco Secondini, Li Zhang, Dario Cellini, Asfand Nizamani, Ramin Solaimani, Lorenzo De Marinis, Mareli Rodigheri, Pantea Nadimi Goki, Muhammad A. Naz, Giampiero Contestabile, Enrico Forestieri, Fabio Cavaliere

Machine Learning

VICT: Verifier-Instrumented Credit Tracing for Long-Horizon LLM Agent Reinforcement Learning

Fine-grained credit assignment is a central challenge in reinforcement learning for long horizon LLM agents. Standard objectives often train from programmatically verifiable terminal rewards by broadcasting each sparse outcome to every action in a trajectory. Existing methods typically seek finer credit from the rollout side, constructing auxiliary trajectory signals or additional comparisons to estimate action importance. Although useful, these approaches still treat the verifier that judged success as a scalar reward, discarding its internal task structure. Our key insight is that many verifiable tasks already encode the relevant checks inside their terminal verifier. We propose VICT (VerifierInstrumented Credit Tracing), a training-time interface that exposes executable or evidence backed atoms and traces them back to actions through dependency-valid proof edges. VICT redistributes group-relative advantage only along those edges, shifting credit assignment from rollout-side inference to verifierside tracing. It preserves the original terminal reward, abstains when evidence is incomplete or ambiguous, and changes only the training-time advantage tensor, requiring no learned critic, process labels, branch rollouts, or inference-time verifier access. On ALFWorld and WebShop, VICT improves substantially over outcome-only training and achieves strong performance alongside recent fine-grained credit methods; ablations rule out dense atom rewards, final-commit credit, temporal proximity, and sparsity as sufficient explanations.
Read the full abstract

All authors Pengcheng Li, Zhengyang Zhang, Dongxu Zhang, Sui Huang, Shaohua Ma

Sound

Exploring the Design Space of Representation Learning for Audio Transformations

Neural audio representation learning has enabled a range of content-oriented applications, but the resulting features remain limited for tasks involving audio processing. Furthermore, it is not obvious what processing-aware representations should capture: the processing itself, abstracted away from source content, or the processed audio that retains it. Existing approaches implicitly commit to one or the other and also differ in their models, data, and evaluation, obscuring which design choices drive their behavior. We address both questions within a unified framework of three objectives: processing consistency, description alignment, and equivariance via forward prediction. We compare all combinations of the objectives under a controlled setup and reveal their relative strengths and interactions. Our framework produces both a transformation embedding and a processed-audio embedding, and we find that the two play complementary roles: distance-based tasks favor the former, while probe-based tasks favor the latter. Combined with improvements in network architecture and training pipeline, our representations outperform prior baselines across retrieval, probe-based evaluation, and style transfer.
Read the full abstract

All authors Sungho Lee, Marco Martínez-Ramírez, Junghyun Koo, Wei-Hsiang Liao, Kyogu Lee, Yuki Mitsufuji

Optics

Self-partitioned Interfacial Time Crystals

Nonequilibrium many-body systems can spontaneously break symmetry in time, as in time crystals, or in space, through self-organized domains and interfaces. Whether these two forms of symmetry breaking can intertwine so that an emergent interface alone hosts time-crystalline order remains unknown. In this work, by introducing the Rabi-Hatano-Nelson model, we unveil the existence and mechanism of a self-partitioned interfacial time crystal (SPITC), where a homogeneous system generates its own internal and tunable interface, at which the time-translation symmetry is also spontaneously broken. Such a SPITC phase is intrinsically induced by nonreciprocity and open boundary conditions, without external pumping or long-range interaction. The periodic and open boundary phase diagrams of the system are both mapped out; vacuum and Dicke-like superradiance with static or active orders are identified, with analytical phase boundaries in the weak coupling limit. The frequency of the SPITC is found to scale quadratically with the spin-photon coupling strength, as we derive analytically for the slow dynamics of the spins. The position of the SPITC boundary scales with a critical exponent of $-1$ as a function of the degree of nonreciprocity, in stark contrast to $-1/2$ for an otherwise stationary boundary. Our construction of SPITC establishes a route to spatiotemporal order in non-Hermitian many-body systems.
Read the full abstract

Mathematical Physics

Explicit Bound States and Threshold Resonances for Two Identical Fermions on a One-Dimensional Lattice

We study a two-particle lattice Schrödinger operator describing two identical fermions on the one-dimensional lattice $\mathbb Z$ with nearest-neighbor interaction of strength $\lambda\in\mathbb R$. Using the direct-integral decomposition with respect to the total quasi-momentum $k\in\mathbb T:=(-\pi,\pi]$, we reduce the problem to fiber operators $H_\lambda(k)$ acting in the odd relative-coordinate space. For $k\in(-\pi,\pi)$, the essential spectrum is \[ \sigma_{\mathrm{ess}}(H_\lambda(k)) = [4-4\cos(k/2),\,4+4\cos(k/2)]. \] We give a complete description of the spectral transition at both edges. The operator has a unique simple eigenvalue outside the essential spectrum if and only if \[ |\lambda|>2\cos(k/2), \] in which case \[ E(k,\lambda) = 4+\lambda+\frac{4\cos^2(k/2)}{\lambda}. \] For $|\lambda|<2\cos(k/2)$ there is no discrete eigenvalue. At the critical coupling $|\lambda|=2\cos(k/2)$, the eigenvalue merges with the corresponding spectral edge and becomes a threshold resonance, with a bounded non-square-integrable odd solution. We also derive the threshold and strong-coupling asymptotics of the eigenvalue and eigenfunction. In the strong-coupling regime the eigenfunction localizes at the interaction sites, whereas at critical coupling it converges pointwise to the corresponding resonant solution. The exceptional fiber $k=\pi$, where the hopping vanishes and the essential spectrum collapses to $\{4\}$, is treated separately.
Read the full abstract

Cryptography and Security

TagZilla: Automated Owner and Abuse Type Tagging for Indicators of Compromise in Threat Reports

Cyber Threat Intelligence (CTI) reports often describe Indicators of Compromise (IoCs) such as IP addresses, URLs, file hashes, and cryptocurrency wallets involved in cyberattacks. Those IoCs are typically described in the unstructured report's text, or listed at the end of the report with little context, limiting their usefulness. This paper presents TagZilla, a platform that, given a threat report, automatically analyzes its text and tags the IoCs it describes with contextual information about the threat group and malware family that the IoC belongs to and the type of abuse associated with the IoC (e.g., phishing, sextortion, command-and-control). TagZilla provides a novel LLM-based approach to assign owner tags to IoCs using an open-world classification, and assigns 29 abuse type tags to IoCs using a closed-world classification. We evaluate TagZilla on a manually generated ground truth of 100 threat reports containing 1,534 indicators, where it achieves an F1 score of 0.94 for owner tagging and 0.93 for abuse type tagging. Then, we apply TagZilla to tag 765 threat reports, identifying 15,583 IoCs belonging to 637 malware families, 113 threat groups, and 162 other entities. The results show that TagZilla can tag IoCs even in reports describing multiple actors and malware families, enabling the generation of IoC profiles for those entities.
Read the full abstract

Fluid Dynamics

A high-order polynomial-corrected shifted boundary method for simulating fully nonlinear water waves

We present a novel unfitted computational framework for simulating fully nonlinear potential flow-based water waves. Focusing on wave propagation, we describe the core methodology, which involves a high-order polynomial-corrected shifted-boundary approximation on unfitted spectral elements. This approach allows for the simulation of a curved, highly time-dependent (moving and deforming) free surface affected by bathymetric changes. All that on a simple Cartesian background mesh without re-meshing or further approximations of non-affine geometric features. In addition, we highlight the importance of proper gradient recovery using a polynomial-preserving technique to accurately capture the vertical free-surface velocity. Ultimately, the goal is to develop a high-order convergent numerical scheme capable of simulating highly nonlinear waves over long periods. This is achieved through an arbitrary-order finite-difference approximation of the free surface combined with added hyperviscosity for numerical stability. We present verification and validation test cases for wave propagation in both periodic and finite domains. Emphasis is placed on convergence studies, the justification for using high-order approximations, the importance of optimal gradient recovery, long-time simulation of highly nonlinear waves, and nonlinear wave interactions with both affine and curved bathymetry changes, as well as with vertical walls, for highly nonlinear stream function waves and high-amplitude solitons. The novel computational model offers a comprehensive, all-in-one framework for simulating ocean waves and their interactions with offshore structures. It provides significant geometric flexibility, enabling boundaries such as the free surface to move and deform over time without the need to re-mesh a boundary-fitted mesh.
Read the full abstract

All authors Jens Visbech, Allan P. Engsig-Karup, Harry B. Bingham, Mario Ricchiuto

Multimedia

Agentic Artifact Creation: Systems, Evaluation, Principles, and Opportunities

Generative models can turn natural-language prompts into images, text, code, and other content, lowering the cost of producing drafts and components. Their practical impact increasingly depends on whether those pieces can become complete, dependable deliverables. This survey examines agentic artifact creation, which we define as stateful construction in which an AI system materially constructs or revises a deliverable and intermediate observations redirect later work. Functionally, the process links an operational representation of the artifact, a construction policy, and runtime verification whose feedback can redirect later actions. We reviewed 259 works available through August 20, 2026: 230 systems meeting this definition and 29 benchmarks of agentic artifact construction. We compare six artifact families, then analyze application settings and evaluation practice as separate dimensions. Across families, construction challenges reflect not only modality but also how tightly decisions are coupled and whether failures become visible while they remain repairable. Decomposition can reduce local complexity while increasing coordination and reassembly costs. Learned judges may add little independent evidence when they share the generator's preferences or blind spots. We formulate principles for keeping commitments and responsibility explicit, turning feedback into targeted repair, and revalidating affected state after change. We also identify opportunities for sustaining coherent, accountable control as artifacts, creator intent, and construction systems evolve. A curated paper list is available at https://github.com/GeminiLight/awesome-agentic-artifact-creation.
Read the full abstract

All authors Tianfu Wang, Zhezheng Hao, Xilin Xia, Lixin Liu, Mengkang Hu, Hongzhang Liu, Xi Chen, Ziyan Liu, Xiankun Lin, Weijia Zhang, Nicholas Jing Yuan, Hui Xiong

Complex Variables

Coefficients and Integral Mean Estimates for $K$-Quasiconformal Harmonic Mappings

Recently, Li and Ponnusamy~\cite{LiPonnusamy2025} established the coefficient conjecture proposed by Wang et al.~\cite{Wang2024} for several prominent geometric subclasses of $\mathcal{S}^0_H(K)$, the class of sense-preserving $K$-quasiconformal univalent harmonic mappings in the unit disk. In this paper, we show that the conjecture continues to hold for a class of $K$-quasiconformal harmonic mappings defined via quasi-subordination. Further, we determine the range of $p>0$ for which such mappings belong to the Hardy space ${\bf h}^p$ and the weighted Bergman space $\mathbf{a}^{\mathbf{p}}_{\boldsymbol{\beta}}$, for $\beta>-1$. Our Hardy space result makes significant progress toward a problem posed by Pavlović, while the Bergman space result sharpens the range obtained by Das and Rasila~\cite{DasRasila}, doubling the previously known bounds. In addition, we obtain refined growth and integral mean estimates for the subclass, improving earlier results and providing further evidence toward an open problem raised by Das et al.~\cite{DasRasila2025}. Parallel results are also discussed for odd $K$-quasiconformal harmonic mappings.
Read the full abstract

Analysis of PDEs

Weighted $L^2$ bounds for kinetic Fokker-Planck equations and hypocoercivity

We study the long-time behaviour of solutions to kinetic Fokker-Planck equations with power law confinement potentials and local equilibria with fat tails. Without relying on a priori moment bounds or perturbative regimes, we establish global estimates on weighted norms (main result) using two methods: a coupling of the weighted norms with entropy dissipation (DMS method) and a Foster-Lyapunov condition approach. As a consequence, we establish an entropy - entropy production inequality and the convergence rates to equilibrium in terms of the exponents associated to the growth of the spatial confinement and the tail decay of local equilibria. For sake of simplicity, we assume that stationary states are factorized.
Read the full abstract

Strongly Correlated Electrons

Field-tuned quasiparticles and electronic structure in heavy-fermion YbNi4P2

We study the Fermi surface topology and quasiparticle properties in the heavy fermion compound YbNi4P2 at high magnetic fields using quantum oscillation measurements. We observe a large decrease of the quasiparticle mass with increasing field and demonstrate good qualitative agreement with the single-ion Kondo model. At the putative Lifshitz transition at 17 T, we observe a sudden change of quantum oscillation frequencies suggesting an abrupt change of the electronic structure and/or quasiparticle characteristics. Our results demonstrate the ability to tune the electronic structure and provide input for theoretical models of YbNi4P2 and correlated electron systems in high magnetic fields.
Read the full abstract

All authors Will Broad, Owen Moulding, Takaki Muramatsu, Manuel Brando, Alix McCollam, Femke Bangman, Gertrud Zwicknagl, Kristin Kliemt, Cornelius Krellner, Sven Friedemann