Peer Revenue Play credits. Real papers.

Papers

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

All of arXiv

Tuesday 8 September

Computation and Language

Compositional Multilingual and Behavioral Attribute Steering

This study examines the compositionality of steering vectors for language and behavioral control in large language models. Focusing on language, jailbreak, and conciseness, we investigate whether additive, training-free composition of attribute steering vectors can preserve the intended steering effect of each attribute, across four instruction-tuned models from two model families and two size scales. We find that single-attribute steering is reliable for all three attributes, but only within an appropriate combination of intervention layer and steering strength, with abstract behaviors (jailbreak, conciseness) favoring middle layers and language favoring earlier layers. We show that additive composition of two attribute vectors succeeds in steering both attributes simultaneously when each is injected at its own best-performing layer, and that this partially extends to three simultaneously composed attributes, addressing an inconsistency left open by prior work on training-free composition. We further analyze the geometric properties of these steering vectors, finding that they are approximately orthogonal in the residual stream, consistent with their compositional behavior.
Read the full abstract

All authors Hyun Gu Kang, Daniil Gurgurov, Tanja Baeumel, Josef van Genabith, Simon Ostermann

Robotics

Coverage Path Planning for Redundant Manipulators using Generalized Spanning Trees

Surface coverage with task-redundant manipulators is challenging because each surface point may admit multiple inverse kinematics (IK) solutions, and configuration choices strongly affect motion quality. This paper extends the classical Spanning Tree Coverage (STC) method to redundant manipulators through offline and online Joint Spanning Tree Coverage (JSTC) algorithms. Offline JSTC samples multiple Inverse Kinematics (IK) solutions per grid cell and formulates the problem as a Generalized Minimum Spanning Tree (GMST), selecting one configuration per cell and tracing the resulting tree to obtain a non-revisiting coverage path. Online JSTC incrementally expands and backtracks a spanning tree with feasibility and cost evaluation while handling dynamic grid updates. Simulation results show that offline JSTC reduces computation time, reconfigurations, and joint motion compared to other methods, while online JSTC achieves fast per-step planning in dynamic scenarios.
Read the full abstract

Robotics

Localized Visual Feature Aggregation via Focus Pooling for Visuomotor Policies

Focusing on spatially localized, control-relevant visual cues has been shown to improve data efficiency in visuomotor policies by reducing the need to model task-irrelevant visual variation. Existing methods often impose this focus through input preprocessing, such as cropping control- or object-centric regions in RGB images or point-clouds. However, it remains underexplored whether such localized features can be exposed directly from commonly used convolutional neural network (CNN) encoded features. In this paper, we show that intermediate CNN features preserve localized visual context for control, but existing pooling methods fail to aggregate it effectively. We introduce FocusPool, an attention pooling module that selectively aggregates intermediate visual features according to their relevance to the robot's current proprioceptive context. The resulting pooled representation captures task-progressive, control-relevant local information and is used directly for policy learning. Across simulation and real-world experiments, FocusPool improves policy success rates over pooling and explicit local focus methods by 36.2% and 41.2%, with training only 5.8% of encoder parameters.
Read the full abstract

All authors Ruiyu Wang, Zheyu Zhuang, Danica Kragic, Florian T. Pokorny

Artificial Intelligence

FastE: Readout-Triggered Token Compression for LLM Embedding Inference

In this study, we identify depth-dependent prefix redundancy in final-readout LLM embedding models, notably across representative backbones including Qwen3-Embedding and Qwen3-VL-Embedding. We find that removing prefix states is substantially more damaging in shallow layers than at greater depth, showing that prefix states become increasingly compressible as the prefix and readout states propagate through the network. To this end, we introduce FastE, a training-free, plug-and-play method. FastE uses a shared fixed threshold on batch-mean readout-prefix alignment as a lightweight online heuristic for selecting when compression occurs, and ranks prefix states by the attention scores they receive from the readout position to determine which states are retained in subsequent layers. Our evaluations demonstrate FastE's ability to substantially reduce computational costs: on NarrativeQA with Qwen3-Embedding-0.6B, it reduces decoder-backbone FLOPs by 40.11% while retaining 99.53% of Full Forward nDCG@10. Across five text embedding benchmarks, two backbone scales, and three cross-modal retrieval tasks, the quality-efficiency trade-off is directly customizable through the maximum removal ratio without retraining. We believe FastE offers practical value for scalable embedding generation in retrieval, indexing, clustering, and multimodal representation systems.
Read the full abstract

All authors Jinsong Shu, Jinyong Wen, Baokun Wang, Zhongle Xie, Lidan Shou, Weiqiang Wang, Gang Chen

Quantum Physics

Performance of the subcarrier-wave quantum key distribution in the presence of spontaneous Raman scattering noise generated by classical DWDM channels

In this paper we study performance of the subcarrier-wave quantum key distribution system (SCW QKD) in the presence of spontaneous Raman scattering (SpRS) noise generated by classical channels of dense wavelength division multiplexing (DWDM) network within a single-mode optical fiber. We present the mathematical model for evaluation of the quantum bit error rate (QBER) and the secure key generation rate with the SpRS noise taken into account. We consider two regimes of the SCW QKD system: the continuous wave regime that uses continuous wave laser and the pulsed regime. For these regimes, performance of the system is analyzed depending on receiver sensitivity of classical DWDM. It is found that the pulsed regime outperforms the continuous wave regime in both the secure key generation rate and the maximum achievable distance.
Read the full abstract

All authors Fedor Kiselev, Roman Goncharov, Natalia Veselkova, Eduard Samsonov, Alexei D. Kiselev, Vladimir Egorov

Systems and Control

On the Local and Global Nature of Frequency and Voltage Dynamics

This paper utilizes the autocorrelation of frequency and voltage measurements to identify, quantify and classify local and global properties of power system dynamics. The analysis is based on measurements with various resolutions (20 ms, 1 s, and 1 min) from several nodes of the Irish All-Island Power System (AIPS). Simulations based on stochastic differential algebraic equations on an IEEE benchmark system support conclusions drawn from real-world data.
Read the full abstract

Machine Learning

Environments as Scaffold: Enriching Feedback to Bootstrap Self-Evolving Agents in Long-Horizon Tasks

Large Language Models demonstrate remarkable proficiency in static reasoning, yet training them as autonomous agents through Reinforcement Learning (RL) for long-horizon tasks is often hindered by severe reward sparsity. While conventional agent-side warming up via supervised fine-tuning (SFT) can alleviate this, it is frequently limited by data scarcity and constrained exploration. To address this, we propose a paradigm shift to environment-side adaptation by constructing Feedback-Enriched Environments (FEEs). Through a pilot study, we establish a feedback design strategy that reformulates environments by transitioning from action guidance to observation enrichment during the later stages of both intra-episode exploration and inter-episode evolution. Large-scale experiments on SciWorld and BFCL benchmarks using various Qwen3 model scales and RL algorithms such as GRPO, GSPO, and DAPO demonstrate that FEEs consistently yield performance improvements over standard settings. Furthermore, our analysis reveals that training with FEEs (1) stabilizes training dynamics by reducing entropy volatility, (2) facilitates proactive state-space exploration in difficult tasks, (3) ensures the internalization of environmental guidance into policy weights rather than acting as a mere inference-time prior, and (4) identifies intra-group feedback consistency as a critical boundary for stable optimization.
Read the full abstract

Statistics Theory

A note on a local entropy condition in the Wasserstein space

Entropy conditions are widely used in empirical-process analyses to establish asymptotic properties of M-estimators. In regression problems with responses taking values in a general metric space, a commonly imposed condition requires the entropy integral associated with a shrinking ball centered at the target object to remain uniformly bounded as the ball radius tends to zero. We show that this condition generally fails in the quadratic Wasserstein space of univariate probability distributions supported on a compact interval. Specifically, when the target object is a strictly increasing and absolutely continuous distribution function whose derivative is bounded away from zero and infinity, the corresponding entropy integral diverges as the ball radius tends to zero. The same divergence persists even when the ambient space is restricted to the class of distribution functions satisfying fixed uniform two-sided Lipschitz bounds. These results indicate that different asymptotic analyses are required for the quadratic Wasserstein space.
Read the full abstract

Computer Vision

Towards Embodied Air-Ground Cooperative Object Search: Benchmark, Dataset and Agentic Method

Air-Ground Object Search (AGOS) in urban environments is a challenging embodied task, which requires an Unmanned Aerial Vehicle (UAV) and an Unmanned Ground Vehicle (UGV) to jointly search for and verify a specified target vehicle from multi-view visual references. To study this underexplored problem, we introduce AGOS-Bench, the first dedicated benchmark for evaluating whether general-purpose Vision-Language Models (VLMs) can integrate aerial discoveries and ground-level verification through UAV-UGV cooperation. We further provide AGOS-Dataset as the companion resource of exemplary trajectories constructed by an automatic pipeline. It consists of 7.7k episodes for searching objects of diverse categories and attributes, spanning three difficulty levels. To address the AGOS task, we propose AGOS-Agent, a training-free and tool-augmented approach. The agentic method relieves VLMs from complex and dynamic coordination via a deliberate search-handoff-verify cooperation protocol, only demanding VLMs for scene understanding and decision-making. Extensive experiments on nine VLMs show that AGOS-Agent improves overall success rate for eight of the nine evaluated backbones while reducing decision steps for all nine. On the hard split, the SR and SPL of Gemini-3.6-Flash increase from 8.6% to 55.7% and from 7.6% to 44.0%, respectively.
Read the full abstract

All authors Boao Yu, Zimo Chen, Junreng Rao, Yue Hu, Zhengqiu Zhu, Yong Zhao, Rusheng Ju

Differential Geometry

The Bimetric generalization of Kastler–Kalau–Walze Type Theorems

Let $M^n$ be a closed oriented manifold of even dimension $n=2m\ge2$, equipped with a smooth metric $g_1$ and a Riemannian metric $g_2$. We compute the Wodzicki residue of $D_{g_1}^2D_{g_2}^{-n}$ on the common exterior bundle, where $D_{g_r}=d+\delta_{g_r}$. This is the noncommutative integral of $D_{g_1}^2$ relative to the reference operator $D_{g_2}$. Its local density involves the curvatures of the two metrics and the difference of their Levi–Civita connections. Integration by parts gives a closed-manifold formula without explicit derivatives of the connection difference. For compact manifolds with boundary and $n\ge4$, we assume that both metrics are Riemannian and satisfy $g_r=h_r(x_n)^{-1}g^{\partial M}+dx_n^2$ near the boundary, with $h_r(0)=1$. We compute the noncommutative residues of two factorizations involving even and odd powers of $D_{g_2}$. Their interior contributions coincide, whereas their boundary terms are explicit multiples of the integral of $K_{g_1}-K_{g_2}$, with the coefficient for the even factorization twice that for the odd one. Here $K_{g_r}$ is the trace of the second fundamental form with respect to the inward unit normal. The calculation uses a second-order residue formula and direct boundary symbol expansions. When the metrics coincide, the formulas reduce to the Hodge–de Rham Kastler–Kalau–Walze identity.
Read the full abstract

Quantum Algebra

Double cross products with projections and relative Rota-Baxter operators on Hopf algebras

Given a matched pair of Hopf algebras, the double cross product construction yields a new Hopf algebra, the prototypical example of which is the generalized quantum double arising from a (Hopf) skew pairing. In this paper, we specifically investigate double cross products with projections in the framework of Radford's biproduct theory, and establish that such an algebraic structure is equivalent to a simplified version of (Yetter-Drinfeld) relative Rota-Baxter operators on Hopf algebras recently introduced by Sciandra, with the corresponding braided Hopf algebras lying in the Yetter-Drinfeld module category under coadjoint coactions. We further examine conditions under which this algebraic structure induces a matched pair of actions on a single Hopf algebra, which is in turn equivalent to the notion of braiding operators on Hopf algebras that yield solutions of the braid equation. In addition, we show that such relative Rota-Baxter operators are morphisms between certain Doi-Hopf modules.
Read the full abstract

Machine Learning

MLIP Detective: Active Failure Mode Discovery Beyond Benchmark Scores for Machine-Learning Interatomic Potentials

Universal machine-learning interatomic potentials (u-MLIPs) aim to generalize across diverse configurations. Benchmarks enable reproducible evaluation but may not expose failures outside their predefined scope. Here, we show that physics-informed search can complement benchmark-based evaluation by uncovering hidden failure modes. We introduce MLIP Detective, an agentic framework for active failure mode discovery. Starting from benchmark evidence, MLIP Detective generates falsifiable, physics-informed failure hypotheses, screens them with inexpensive simulations, and escalates only the most suspicious cases to human experts together with proposed verification protocols. Without issue-specific prompting, MLIP Detective identified and characterized a systematic anomaly in MACE-MPA-0: the model predicted some relaxed adsorbate-surface systems involving O- or F-containing adsorbates to be higher in energy than their corresponding separated fragments. Using cross-model comparisons, MLIP Detective further inferred a likely training-data origin for the anomaly, consistent with recent reports.
Read the full abstract

All authors Ryuhei Okuno, Nontawat Charoenphakdee, Kaoru Hisama, Yuta Tsuboi

Number Theory

Prescribed Abscissae on Congruent-Number Curves over Simplest Cubic Fields

For integers $t\geq -1$, let $\theta_t$ be the largest real root of the Shanks polynomial $g_t(X)=X^3-tX^2-(t+3)X-1$ and put $K_t=\mathbb{Q}(\theta_t)$. We classify the points on $E_n:y^2=x^3-n^2x$ over $K_t$ with abscissa $\theta_t-1$ when $n$ is a positive integer and $E_n(\mathbb{Q})$ has rank zero. We determine the four pairs $(d,t)$ with $d>0$ rational for which $(\theta_t-1)/d^2$ is an abscissa on $E_3(K_t)$. We also exclude the abscissa $\theta_t-1$ on $E_5(K_t)$ and $E_6(K_t)$ and prove that only finitely many parameters $t$ admit this abscissa for each fixed positive integer $n$.
Read the full abstract

Numerical Analysis

A Discontinuous Galerkin Method for the Intrinsic Beam Model with Kelvin-Voigt Damping

We present an energy-stable discontinuous Galerkin (DG) spatial discretization of the intrinsic beam equations with Kelvin-Voigt damping. We reformulate the governing equations by splitting the total sectional resultants into elastic and viscous parts, thereby exposing a mixed system without mixed space-time derivatives. For homogeneous cantilever boundary data, the continuous system obeys an energy dissipation identity. With characteristic upwinding and alternating auxiliary traces the DG discretization is energy stable at the semidiscrete level for the stated homogeneous split boundary closure. The method is implemented in the Julia-based simulation framework Trixi.jl. A physically compatible manufactured solution with all state and viscous components active assesses convergence, while a rotating-beam test verifies the analytic constant-spin branch and the actuated energy balance.
Read the full abstract

All authors Shivam Sundriyal, Christian Bleffert, Lukas Dreyer, Melven Röhrig-Zöllner, Gregor Gassner, Vadym Aizinger

Astrophysics of Galaxies

The Impact of Circumgalactic Rotation on Ly$\alpha$ Radiative Transfer

Hydrogen Lyman-alpha (Ly$\alpha$) is a prominent emission line from the circumgalactic medium (CGM). Due to its resonant nature, Ly$\alpha$ carries imprints of the physical properties and kinematics of the cold CGM. In particular, CGM rotation can modify the Ly$\alpha$ peak separation, which is often interpreted as a tracer of H I column density. We present 3D Monte Carlo Ly$\alpha$ radiative-transfer simulations in a CGM-like rotating medium and examine how the emergent spectra depend on rotational velocity ($V_{\rm rot}$), H I column density ($N_{\rm HI}$), viewing angle, clumpiness, and intrinsic source width. We find that rotation broadens the integrated spectra and increases the peak separation, with the strongest viewing-angle dependence when rotational Doppler shifts dominate over frequency diffusion. At high $N_{\rm HI}$, numerous scatterings reduce the sensitivity of integrated spectra to rotation, producing a degeneracy between $V_{\rm rot}$ and $N_{\rm HI}$. Consequently, Ly$\alpha$ peak separation alone can overestimate $N_{\rm HI}$ in a rotating medium. Spatially resolved halo spectra provide a clearer diagnostic: opposite sides of the rotating medium show systematic redshifted and blueshifted asymmetries associated with the line-of-sight velocity of the last-scattering gas. Such rotation-driven signatures can also contribute to velocity-map patterns often interpreted in terms of inflow or outflow, highlighting the need to consider rotation in spatially resolved Ly$\alpha$ observations. We further show that the main signatures persist in simple clumpy media, while the halo signatures are largely insensitive to the intrinsic source width. Our results demonstrate that spatially resolved Ly$\alpha$ observation is essential for disentangling CGM rotation from radiative-transfer effects.
Read the full abstract

All authors Hee-Gyeong Kym, Seok-Jun Chang, Max Gronke, Kwang-Il Seon

Probability

Planarity and number of crossings in general models of random geometric graphs

Consider a random geometric graph with vertices given by a Poisson point process, and whose edges depend on independent marks corresponding to the vertices and pairs of vertices. In this paper, we study two related questions on this general model: the number of edge crossings in a projection of this graph, and its graph-theoretical planarity. We focus on models with heavy-tailed mark distributions, in particular with polynomial tails with arbitrary exponents. We show that the asymptotic behaviour varies significantly depending on this exponent.
Read the full abstract

Cryptography and Security

Windows Malware Detector as a Compound AI System: Trade-Offs in Accuracy, Efficiency, and Adversarial Robustness

Industrial Windows malware detectors are commonly described as Compound AI Systems composed of multiple heterogeneous components, including rule-based mechanisms as well as machine-learning-based static and dynamic analyses. However, due to industrial secrecy and limited public disclosure, the internal architectures of these systems can only be inferred, rendering systematic evaluations of detection accuracy, computational costs, and adversarial robustness largely infeasible. In contrast, academic research provides reproducible and transparent evaluation methodologies, but typically investigates individual detection components in isolation. To bridge the gap between academic research and industrial practice, and inspired by state-of-the-art industrial architectures for Windows malware detection, we propose a novel methodology that (i) explicitly balances the trade-off among detection performance, computational requirements, and robustness, and introduces (ii) system-level threat models that capture how attackers exploit different degrees of knowledge to evade the entire Compound AI System rather than isolated detectors. Experiments conducted on real-world data demonstrate that the Compound AI System training time can be reduced and responsiveness improved while incurring only a marginal loss in detection performance. Leveraging our threat modeling, we show that increasingly knowledgeable attackers craft more effective adversarial examples, revealing the system's strengths and weaknesses, degrading its responsiveness, and exposing a direct trade-off between efficiency and robustness. Finally, we translate these trade-offs into take-home messages and deployment guidelines, helping practitioners to select the system that best matches their operational constraints.
Read the full abstract

All authors Andrea Ponte, Luca Demetrio, Luca Oneto, Battista Biggio, Fabio Roli

High Energy Physics (Phenomenology)

Light cone distributions of $P$-wave quarkonia

Motivated by renewed interest in the light-cone distributions of $P$-wave quarkonia, we investigate the leading-twist distribution amplitudes of these states within the basis light-front quantization (BLFQ) formalism. While our extracted distributions exhibit macroscopic shapes consistent with expectations from the non-relativistic limit, we find that relativistic effects introduce critical structural features. Most notably, we observe novel “W"-shaped structures in the distribution amplitudes of the axial vector mesons, which arise from relativistically induced $S/D$ partial waves and cannot be explained by non-relativistic dynamics. These findings provide essential non-perturbative inputs for analyzing hard exclusive processes and highlight the importance of relativistic frameworks for understanding heavy quarkonium production and structure at modern high-energy colliders.
Read the full abstract

All authors Yang Li, Tianyang Hu, Xianghui Cao, Meijian Li

High Energy Physics (Experiment)

Prospects for precision Higgs boson measurements at a 10 TeV muon collider

A 10 TeV muon collider offers a powerful environment for precision studies of the Higgs sector through its large vector-boson-fusion production rates. This review presents projected sensitivities to single- and double-Higgs production cross sections, the Higgs boson mass, and the trilinear Higgs self-coupling. The projections are obtained from detailed simulations of the MUSIC detector, including machine-induced background, for a baseline configuration of two experiments, each collecting an integrated luminosity of 10 $\mathrm{ab}^{-1}$. The $H\to b\bar{b}$ and $H\to WW^\ast$ production cross sections can be measured with statistical precisions of 0.18% and 0.35%, respectively, while precisions of 2.6%, 4.2%, and 6.9% are expected for $H\to\gamma\gamma$, $H\to ZZ^\ast$, and $H\to\mu^+\mu^-$. Combining the $H\to b\bar{b}$, $H\to\gamma\gamma$, and $H\to\mu^+\mu^-$ channels yields an expected Higgs boson mass precision of about 19 MeV. Double-Higgs production can be measured with statistical precisions of 4.2% in the $HH\to b\bar{b}b\bar{b}$ channel and 14% in the $HH\to b\bar{b}WW^\ast$ channel. Using the $HH\to b\bar{b}b\bar{b}$ channel, the trilinear Higgs self-coupling can be determined with an expected precision of about 5%. These results demonstrate the potential of a high-energy muon collider for precision Higgs physics.
Read the full abstract

All authors Paolo Andreetto, Massimo Casarsa, Luca Castelli, Alessio Gianelle, Donatella Lucchesi, Leonardo Palombini, Imran Raghib, Lorenzo Sestini, Davide Zuliani

Computer Vision

From Coordinates to Candidate Regions: Temporal Change Localization via Region Selection in Remote Sensing Multimodal LLMs

Remote sensing multimodal large language models (RS-MLLMs) have advanced scene understanding and visual question answering over satellite imagery, yet localizing specific objects or changed regions remains challenging. Existing approaches rely on generating bounding box coordinates as token sequences, which is fragile for the small, densely packed objects common in remote sensing and increasingly error-prone when multiple targets must be localized simultaneously. In this work, we present an RS-specific formulation of the region selection paradigm, previously explored in natural-image MLLMs, and extend it to temporal change localization over multi-image sequences. Our framework employs a text-conditioned region proposal module, encodes each candidate as special tokens carrying per-frame visual features enriched with spatial and temporal cues, and lets the LLM localize targets by selecting region tokens in its response. We construct a multi-task training and evaluation suite spanning localization, referring expression, visual grounding, and understanding tasks across single-image and multi-temporal settings. Experiments show that our approach substantially outperforms coordinate-generation baselines on temporal change localization, while improving single-image visual grounding and maintaining competitive understanding performance. Oracle analysis decomposes the contributions of the region proposer and the LLM selector, providing diagnostic insight unique to this framework. Our code will be available at https://github.com/juwan-kr/RS-RegionSelect.
Read the full abstract

All authors Juwan Chung, Sungjune Park, Yeongyun Kim, Yong Man Ro

Sound

Noise Adaptive Streaming Audio-Visual Speech Token Enhancement for Robust Full-Duplex Spoken Dialogue Models

Full-duplex spoken dialogue systems enable simultaneous listening and speaking, but their audio-only perception often fails under background noise and overlapping speech, leading to incoherent responses. Recent audio-visual dialogue approaches show that incorporating visual cues such as lip movements improve robustness under audio corruption. However, existing approaches often adapt the large speech dialogue model itself to process visual input, requiring costly multimodal training. We propose AV-STE, a modular streaming audio-visual front-end that restores corrupted semantic speech tokens from noisy audio and lip video before they reach the speech LLM. The downstream dialogue model remains entirely frozen, preserving its pretrained conversational capabilities. When integrated with frozen Moshi, AV-STE improves average GPT-4o-judged response coherence from 1.42 to 1.91 under same-dataset speaker interference while largely preserving turn-taking behavior. Gains also transfer to out-of-domain Seamless Interaction.
Read the full abstract

Quantum Physics

A theoretical study of subcarrier-wave quantum key distribution system integration with an optical transport network utilizing dense wavelength division multiplexing

In this paper we study the performance of subcarrier-wave quantum key distribution (SCW QKD) in the presence of classical channels of optical transport network utilizing dense wavelength division multiplexing (DWDM). We consider the impact of spontaneous Raman scattering noise as well as the four-wave mixing and channel isolation efficiency. We calculate quantum bit error rate as well as the secure key generation rate of SCW-QKD protocol for different parameters of DWDM system and quantum channel allocations. Our calculations show, that quantum channel allocation at the wavelength of 1310 nm is preferable to allocation at C-band due to lower nonlinear noise, despite fiber's larger attenuation at this wavelength.
Read the full abstract

All authors Fedor Kiselev, Natalia Veselkova, Roman Goncharov, Vladimir Egorov

Quantum Physics

Dynamics of entanglement in non-Hermitian system of nonreciprocal coupling

The exploration of the effect of non-Hermitian (NH) in quantum systems is gaining renewed momentum as a result of the recent progress in experiments. It has been indicated that the ability to engineer the orders of exceptional points (EPs) can be fundamental to its applications in quantum control, such as accelerating entanglement generation; however, an initial survey of the interactions between different orders of EPs is still missing. In this work, we try to partially address this issue by employing a non-reciprocally coupled cavity system which can be experimentally realizable with a two-dimensional reconfigurable NH gauged laser array. With deliberate design, our systems display both second and third orders of EPs. The sign of a novel, purely quantum continuous entanglement phase transition is observed, which reveals how multi-EP interactions alter entanglement dynamics across different regimes while maintaining accelerated generation. Our findings reveil the genuinely quantum NH physics of higher-order EPs, which has opened up a new avenue for future investigations of the higher-order degeneracy application and the dynamic phase transitions of quantum systems.
Read the full abstract

All authors Jiguo Wu, Shaojie Yang, Xiong Liu, Zhibo Jiang, Zhiyou Zhang, Rongchun Ge

Computational Engineering

Risk-Aware Generative Inpainting for Optimized Design Editing of EV Battery Cooling Channels

Cooling-channel layouts for electric-vehicle battery packs must deliver temperature uniformity and low pressure drop while maintaining a single continuous channel. In late-stage design, local modification is a practical way to improve performance while retaining established global features. Diffusion-based inpainting supports such local edits, but its stochastic nature produces different outcomes even for the same region. This raises a fundamental question: how should edit locations be selected when each modification's outcome is stochastic? We propose a risk-aware generative editing framework that selects edit locations by accounting for this variability. Rather than scoring each candidate location by a single expected improvement, the method estimates a distribution of outcomes from offline edit results evaluated with a CFD-trained surrogate, and ranks locations by a chosen risk level. A single trained model therefore supports different editing preferences at inference time, emphasizing either higher expected improvement or greater consistency, without retraining. The policy is evaluated against random editing in a held-out paired study across seven mask configurations. It improves the cooling-channel objective over random editing in most configurations, and the advantage holds under independent CFD verification of the edited designs. Varying the risk level reveals a consistent trade-off between mean improvement and run-to-run consistency, while the learned distribution is useful for ranking locations but should not be read as a calibrated probability distribution. Together, these results show that stochastic generative editing can be converted from a source of variability into a controllable design decision through risk-aware location selection: effective editing depends not only on how a design is modified, but also on where and how much outcome variability is acceptable.
Read the full abstract

Human-Computer Interaction

"Here Be Sharks!": Enhancing Scientific Communication and Analysis through Authoring Interactivity

We report on a case study for designing authoring environments for interactive visualizations to enhance scientific work. We conducted a workshop and prototype review with a group of marine biologists. When it came to visualizing their data, participants identified challenges in conveying their research accurately and completely as well as and analyzing it with ease. Based on our findings, authoring environments should consider the context of scientific work aspects such as domain expertise, collaboration culture, and publication traditions. We propose consideration of non-traditional programming languages and environments for scientific work, discuss ways of facilitating interactive visualizations for scientists, and examine ways of meaningfully integrating AI. The critiques, artifacts, and reactions from the scientists along with our analysis and discussion inform how computational tools should be designed for scientific work.
Read the full abstract

All authors Caroline Berger, Josh Pollock, Dylan Wooton, Arvind Satyanarayan, Clemens Nylandsted Klokmose

Computer Vision

GALoc: Gravity Aligned Wireframes for Depth-Free Monocular Floorplan Localization

Floorplans are compact, appearance-invariant maps ideal for indoor localization, yet existing methods rely on depth networks that are brittle in cluttered scenes. We propose GALoc, a geometry-first framework that replaces depth prediction with gravity-aligned wireframes that satisfy verticality and coplanarity by construction. Given monocular RGB, camera intrinsics, relative poses, and IMU orientation, GALoc constructs a linear constraint matrix encoding verticality and coplanarity, and finds the camera gauge minimizing its smallest singular value via global search. The rectified wireframes are projected into bird's-eye-view layouts through a closed-form, FOV-consistent transformation and matched against the floorplan via metric-free SE(2) search. We evaluate end-to-end on Structured3D, with calibrated noise on Gibson, and on real-world author-collected sequences. When sufficient wall geometry is visible, GALoc matches or outperforms depth-based baselines – achieving 88% sequential localization success at 0.1m over 100-step sequences on Gibson vs the baseline's 68% – while abstaining in structure-blind scenes.
Read the full abstract

All authors Jeahn Han, Minji Kim, Jeongbin Sohn, Jonghyeok Park, Matthias Wuest, Pyojin Kim

Optics

BRAHMS: A cross-platform graphical toolkit for (3+1)D simulation of three-wave mixing in $\chi^{(2)}$ nonlinear media, with GPU and CPU backends

We present BRAHMS, a cross-platform (Linux/Windows) graphical user interface, with GPU (CUDA) and CPU (OpenMP) backends, for the efficient and accurate simulation of three-wave mixing processes involving focused and pulsed Gaussian beams. The package solves the coupled $\chi^{(2)}$ nonlinear Schrödinger equations, including diffraction, dispersion, walk-off, and phase-mismatch effects simultaneously, in full (3+1)D. To our knowledge, this is the first open-source package that solves the complete second-order (three-wave-mixing) nonlinear-optics problem in (3+1)D while fully exploiting the parallel capabilities of modern GPUs, with an equivalent CPU backend available for users without GPU hardware. The GPU implementation is inherently scalable thanks to its Thrust-based design. The package provides a valuable tool for experimental design and for studying three-dimensional field propagation in nonlinear three-wave interactions, supporting applications such as second-harmonic generation (SHG), sum-frequency generation (SFG), and optical parametric generation (OPG). Its GUI requires no programming experience, so that users can readily set up simulations and interpret their results.
Read the full abstract

Atmospheric and Oceanic Physics

GCMagicc v1: a fast generative emulator for multivariate climate-impact ensembles

Projecting the impacts of climate change requires large ensembles of climate variables that match historical observations, align with the warming ranges assessed by the IPCC, and can efficiently run new future emissions scenarios, including the newest generation of climate model scenarios (CMIP7) and pathways consistent with countries' Paris Agreement pledges. Generating such ensembles at the scale needed for impact studies is normally computationally prohibitive. We close this gap with GCMagicc, a hybrid model that pairs a simple physical climate model with machine learning to generate ensembles of 10 climate variables at the resolution of full-scale Earth system models, without relying on GPU resources or retraining for new scenarios. Trained on 32 CMIP6 Earth system models and observational/reanalysis data, GCMagicc complements rather than replaces Earth system models. We apply it to a range of future pathways: the canonical SSP scenarios of the latest IPCC report (1.2-6.1{\deg}C warming, min-max across scenarios of 5-95 percentile ranges), current policies (2.3-4.0{\deg}C), national pledges under the Paris Agreement (1.5-3.3{\deg}C) and the CMIP7 range from the 'VL' to 'H' scenarios (1.2-4.2{\deg}C), releasing a large public dataset. As an illustration, we perform an attribution analysis of the severe 2025 Iranian drought using GCMagicc ensembles, with three CMIP6 large ensembles for comparison, with and without anthropogenic forcings. The results suggest a strong anthropogenic signal: a median probability of drought at least as severe as observed of 29% with anthropogenic forcing, and zero under natural-forcing-only simulations. In the future, drought conditions are projected to materially worsen, amplifying the potential for agricultural and food security impacts and geopolitical conflicts that use water scarcity as a weapon. GCMagicc data is available at https://gcmagicc.org.
Read the full abstract

All authors Nicolai Meinshausen, Malte Meinshausen, Jared Lewis, Zebedee Nicholls, Sarah Schöngart, Alister Self, Xinwei Shen, Karla Spiller, Elisabeth Vogel

Quantum Physics

Non-Markovian dynamics of a qubit on the zigzag edge of photonic graphene

We present an analytical study of the dynamics of a qubit coupled to the zigzag edge of a photonic honeycomb lattice. Leveraging a resolvent operator approach and exact lattice Green's functions, we map the spectrum of qubit-photon bound states and complex poles to uncover the long-time scaling laws of the qubit population. Unlike a qubit coupled to the bulk of graphene, the presence of a zero-energy flat band connecting the Dirac points suppresses long-time branch-cut-induced tails, instead driving damped Rabi oscillations. Potential experimental platforms are briefly discussed.
Read the full abstract

Machine Learning

Equivariance Breaks the Learning Rate

Equivariant networks are commonly trained with Adam, yet recent work reports that matrix structured optimizers such as Muon can perform better, with the reasons for these gains only partly understood. We identify one source of this difference inside equivariant layers. An equivariant layer learns one channel mixing matrix $W_l$ per degree $l$, which we call an irrep block, and shares it across the $2l+1$ components, giving the expanded map $W_l \otimes I_{2l+1}$. This sharing sums gradient contributions across components and can produce different update scales under SGD. Adam's entrywise normalization reduces sensitivity to gradient scale, but neither optimizer directly controls the effective step size of each block. A single learning rate can therefore produce different effective step sizes across blocks. Muon instead controls the effective step size by approximately equalizing the singular values of each momentum matrix. We normalize each irrep block update by a single scalar, preserving its singular value ratios while letting the learning rate control its size. We implement this with spectral normalization or a simpler root-mean-square normalization. We evaluate spectral normalization in a controlled $\mathrm{SO}(3)$-equivariant model with a matched non-equivariant model. In this setting, the step size mismatch grows with width in the equivariant model but not in the non-equivariant model. We evaluate both variants across molecular force prediction on the rMD17 and MD22 datasets, QM9 molecular property prediction, and charged particle dynamics. Across these applications, block normalization generally improves Adam and closes part of its gap to Muon. These results highlight an overlooked interaction between equivariant architectures and their optimizers. Studying and designing the two together may help explain and address training difficulties often attributed to equivariance itself.
Read the full abstract