Soft Condensed Matter
Hydrodynamics of perceptual matter: from neural representations to collective motion
Mohammad Salahshour
How do neural representations of neighbors become collective material behavior? Harmonic Theory (HT) reduces bearing representations to signed perceptual channels traceable to sensory acuity, decision dynamics, and interaction distance. Here we derive a nonlocal kinetic and hydrodynamic theory from the HT particle law without introducing heading
alignment. The resulting fields retain HT's order-resolved perceptual spectrum and keep the neural and sensory origin of each coefficient visible. The theory predicts a torque-driven density instability: HT's direct-approach channel turns
agents toward density excesses, while self-propulsion carries them inward. For isotropic spatial interactions and positive rotational
diffusion, perceptual channel $n$ generically first affects conserved density at spatial order $k^{2n}$, and the full radial transforms select finite pattern wavelengths. Two populations can share the same long-wave instability threshold yet select different initial pattern spacings because their higher perceptual channels differ. Beyond onset, the one-particle continuum sustains a bidirectional nematic column; a prescribed,
symmetry-allowed two-stream response amplifies a directional imbalance into sustained polar flow. In a distinct pure-repulsion regime, one-particle calculations show polar transport and translating transverse bands, with late pattern selection sensitive to numerical transport. The framework connects neural and sensory mechanisms to continuum coefficients, instabilities, and
alignment-free collective motion.
Read the full abstract ▾
Artificial Intelligence
From Papers to Mechanisms: An Evidence-Grounded Knowledge Substrate for Scientific Language Models
Qiuhui Chen, Yibo Liu, Tao Dai et al. (+3 more)
Scientific
language models often access literature through untyped text chunks, which fragment the functional and evidential structure required for mechanism-rich questions. We introduce an evidence-grounded mechanism knowledge substrate that organizes scientific literature into provenance-linked evidence units, role-typed entities, and directed mechanism paths. We instantiate it as MS$^3$, a Material-Sensor-Signal-System schema for conductive-fiber flexible sensors, over 13,689 papers, 131,083 evidence items, and 26,648 mechanism objects. On in-domain and coverage-shift question-answering
benchmarks, we compare closed-book generation, Web search, Raw-PDF RAG, and MS$^3$
retrieval across ten
language models. MS$^3$ improves macro-averaged scientific correctness. It also improves citation entailment and answer completeness. These results support mechanism substrates as a reliable representation layer for scientific
language models and motivate a source-repair workflow in which insufficient MS$^3$ evidence triggers targeted
retrieval from its linked papers rather than assuming that a user has already supplied the correct PDFs.
Read the full abstract ▾
All authors Qiuhui Chen, Yibo Liu, Tao Dai, Jiafan Lu, Zhenglei Zhou, Weimin Zhong
Quantum Physics
High-Fidelity Remote Graph State Preparation for Blind Quantum Computation
Jiawei Cai, Rex Fleur, Benedikt Tissot et al. (+2 more)
Measurement-based
quantum computation (MBQC) relies on entangled graph states, yet existing remote state preparation (RSP) protocols prepare only separable states, requiring subsequent entangling gates on the remote server. Here, we introduce Remote Graph State Preparation (RGSP), a framework that prepares arbitrary graph states directly from a single high-dimensional photonic qudit. By encoding multiple
qubits and their graph connectivity into the
photon's structured phase profile, RGSP can reduce or, for small computations, completely eliminate the need for server-side entangling operations among
qubits. We show that under a reasonable noise model the resulting state fidelity is independent of the graph
topology. Under cumulative fiber phase drift, we demonstrate that this fidelity is significantly enhanced by a “highest-weight-first” mode reordering strategy. Finally, we show that RGSP enables reduced-SWAP universal blind
quantum computing on arbitrary
topologies, reducing
qubit overhead e.g., for a six-
qubit Quantum Fourier Transform from $420$ RSP
qubits with a standard brickwork
topology to $48$ RSP
qubits. These results establish RGSP as a resource-efficient,
topology-invariant primitive for
quantum-secured cloud computing.
Read the full abstract ▾
All authors Jiawei Cai, Rex Fleur, Benedikt Tissot, Wolfgang Löffler, Tzula B. Propp
Combinatorics
On rigidity properties of unit-distance graphs
Sean Dewar, Georg Grasegger, Alison La Porta et al. (+2 more)
A unit-distance graph is a graph which admits a realisation in Euclidean space in which every edge has unit length. Imposing further
geometric conditions on the non-edges gives a family of natural subclasses. Requiring that no two vertices lie at distance less than one gives penny and marble graphs: the contact graphs of collections of equal radii $d$-dimensional spheres with non-overlapping interiors for $d=2,3$. Requiring instead that the straight-line drawing in the plane be non-crossing gives matchstick graphs. Since any motion of a realisation must preserve these extra conditions, the
rigidity and flexibility properties of the resulting frameworks differ from those of classical bar-joint
rigidity theory. In this note we analyse various
rigidity problems for penny and marble graphs, matchstick graphs and unit-distance graphs. In particular we answer a recent open problem on penny and marble graph
rigidity and establish a link between penny graphs and the concept of NAC-colourings.
Read the full abstract ▾
All authors Sean Dewar, Georg Grasegger, Alison La Porta, Jan Legerský, Anthony Nixon
Quantum Physics
Probing Quantum Anomalous Hall Transport Under Microwave Irradiation Using a Topological Circulator
Athul Ashok, Frank Jin, Nick Du et al. (+7 more)
Edge magnetoplasmons (EMPs) provide a platform for probing
chiral charge dynamics and nonreciprocal microwave transport in topological
quantum materials. However, detecting small perturbations to EMP propagation remains challenging because their signatures in conventional microwave scattering measurements can be weak. Here, we investigate microwave-
photon-induced perturbations of EMP transport using a
quantum anomalous
Hall topological circulator. Pump–probe measurements reveal a strongly frequency-selective response: while microwave irradiation substantially modifies the EMP transmission at several pump frequencies, the transmission remains nearly unchanged at selected frequencies, including 4 and 7 GHz. We further exploit the non-Hermitian mode hybridization of the coupled EMP-resonator system to probe the response near an exceptional point (EP). The pump-induced change in transmission magnitude near the EP is approximately twice that observed away from the EP, demonstrating an enhanced microwave response to perturbations. These results reveal the interplay among
chiral EMP transport, microwave-
photon-induced perturbations, and non-Hermitian dynamics, and demonstrate the potential of topological circulators as a promising platform for enhanced microwave spectroscopy and sensing.
Read the full abstract ▾
All authors Athul Ashok, Frank Jin, Nick Du, Luis A. Martinez, Jenny Zhou, Sean O'Kelley, Zachary J.-R. Espley, Gang Qiu, Kang L. Wang, Dong-Xia Qu
Machine Learning
Constrained Goal-directed Planar Graph Generation with Grammar-based Reinforcement Learning
Nicolas Hochuli, Lorenzo Miele, Kristina Shea et al. (+1 more)
Planar graphs are central to applications across science and engineering, yet existing generators provide limited support for goal-directed generation under hard structural and
geometric feasibility constraints. We propose a
dataset-free method for generating planar graph
embeddings by combining parametric graph grammars with safe reinforcement learning to optimize generic task-specific objectives while satisfying constraints during construction. We formulate the generation process as a constrained Markov decision process, where the graph grammar defines the state and action spaces. We further introduce an action projection that maps sampled actions toward state-dependent safe sets, improving constraint satisfaction during training. In contrast to classical graph generators and deep generative models, which typically offer limited goal-directed control or rely on weak constraint satisfaction, our method constructs feasible planar graph
embeddings directly during generation. We also introduce a
benchmark suite for constrained and goal-directed planar graph generation, together with classical and deep generative baselines. Across all
benchmark tasks, our method consistently outperforms baselines while satisfying the formulated constraints.
Read the full abstract ▾
All authors Nicolas Hochuli, Lorenzo Miele, Kristina Shea, Tino Stankovic
Machine Learning
RoSA: Rotational Sparse Adaptation for Memory-Efficient Fine-Tuning
Muhammad Azeem Lodhi, Chao Zhou and Rebekka Burkholz
Parameter-efficient fine-tuning (PEFT) reduces the cost of adapting foundation models by focusing training on a small parameter subset. Complementary to this idea, we introduce RoSA (Rotational Sparse Adaptation), which narrows adaptation to a subset of layers at a time. RoSA freezes lower layers close to the input throughout training and rotates a trainable block over later layers, progressively increasing the number of frozen layers close to the input. This design reduces optimizer-state
memory, shortens backpropagation, and even forward propagation if activations at the last frozen layer are cached. Because RoSA is orthogonal to the choice of trainable parameterization, it can be combined with PEFT methods or sparse optimizers within each active block. Experiments across multiple LLM architectures and tasks show that RoSA reduces peak
memory while maintaining strong fine-tuning performance.
Read the full abstract ▾
Atomic Physics
The influence of the binding energy on Compton-scattering-driven ionization
N. Melzer, M. Kircher, G. Kastirke et al. (+31 more)
We present a comprehensive study of Compton scattering of 20 keV
photons by bound electrons over a wide range of binding energies (Ne 2p, Ne 2s, C 1s, O 1s, and Ne 1s orbitals). In particular, we have measured fully differential cross sections, i.e., electron angular distributions with respect to the direction of the
photon momentum transfer for fixed electron energy. Our data show strong deviations from the predictions of the ”quasi-free” electron approximation. The binding energy and the shape of the Coulomb potential from which the electron escapes both play a significant role in its kinematics. We find that Compton electrons are scattered and even back-reflected by the ionic potential. The strength of this effect depends on the orbital from which the electron is ejected.
Read the full abstract ▾
All authors N. Melzer, M. Kircher, G. Kastirke, J. Rist, D. Tsitsonis, A. Pier, L. Kaiser, J. Kruse, N. Anders, J. Stindl, L. Sommerlad, O. D. McGinnis, M. Schmidt, L. Nowak, A. Kügler, I. Dwojak, P. Daum, P. Roth, M. Astaschov, J. Drnec, K. Lin, N. Wong, F. Trinter, M. S. Schöffler, L. Ph. H. Schmidt, D. M. Haubenreißer, N. M. Novikovskiy, N. Diehl, Ph. V. Demekhin, S. Brennecke, M. Lein, S. Eckart, T. Jahnke, R. Dörner
Machine Learning
Few-Shot Prototype Head Adaptation for On-Device ECG Personalization on PSoC~6
Guilherme Silva, Pedro Silva, Gladston Moreira et al. (+1 more)
Wearable and bedside electrocardiogram (ECG) monitors must adapt to patient-specific morphology to maintain arrhythmia detection accuracy across users, yet personalization is typically performed offline and cannot account for individual physiology, electrode placement, or recording drift. On-device adaptation by backpropagation is expensive for microcontroller-class medical devices because it requires an optimizer state, repeated backward passes through convolutional layers, and labeled arrhythmic beats that may not be available at deployment time. This letter proposes prototype-only head adaptation as a compact personalization primitive for TinyML ECG systems. A one-dimensional convolutional neural network (1-D CNN; 1,314 parameters and 72.6k multiply-accumulate operations per beat) is trained offline on the MIT-BIH Arrhythmia Database under an inter-patient protocol, frozen as a feature extractor, and exported to a PSoC 6 microcontroller. Patient-specific adaptation then reduces to computing closed-form class means in a 32-dimensional
embedding space, requiring no convolutional backward pass, no iterative optimization, and only one forward pass per support beat. Prototype adaptation improves inter-patient macro-F1 from 0.635/0.639/0.646 to 0.731/0.771/0.797 at 1/5/10-shot, outperforming linear
stochastic-gradient-descent (SGD) head fine-tuning at every shot count for the target tiny backbone. On-device replay over 18 one-shot episodes on a PSoC 6 Cortex-M4F matches the host macro-F1 for the prototype head (0.798), with 11.39 ms per beat, 5.2 KB flash, and 22.2 KB SRAM. A restricted variant that updates only the normal-class prototype from passively buffered sinus beats yields a consistent +0.05 macro-F1 gain, reducing the annotation burden during initial
Read the full abstract ▾
All authors Guilherme Silva, Pedro Silva, Gladston Moreira, Eduardo Luz
Computers and Society
Artificial Intelligence and the New Science of Culture
Douglas R. Guilbeault, Bhargav Srinivasa Desikan and James A. Evans
Cognition and culture have long been treated as parallel objects of study, joined more by metaphor than by mechanism. We argue they are co-constituted in a manner now empirically tractable: subjectivities can be measured as local trajectories through a high-dimensional cultural field, and the field is itself the aggregate of those trajectories. Recent advances in machine learning, including
embedding methods and large generative models, provide the first general framework for measuring this co-constitution from micro-cognition to macro-social structure. Vector geometry recovers individual conceptual structure, organizational communication, and large-scale ideologies; captures
multimodal cultural content beyond text; and generates testable predictions about cultural emergence, including the simultaneous arrival of independent discoveries across distant minds. We treat this predictability of simultaneous innovation as evidence for the co- constitutive view: when the geometry of the cultural field is measurable, trajectories of cognitive search through it become predictable. We then examine generative AI
agents as simulators of human subjectivity and as a qualitatively new coordinating substrate whose insertion into social life introduces evolutionary dynamics unprecedented in human cultural history. We formalize this reflexive condition as a Heisenberg-like Uncertainty Principle for generative AI: as instruments for fixing a culture's position grow more precise, our capacity to predict its trajectory degrades, because the machinery of cultural description and cultural action have merged. We close with ethical questions raised by AI-mediated cultural drift, recursive synthetic culture, and the limits of human cultural agency, offering guidelines for safe, equitable, and
privacy-preserving research on AI and culture.
Read the full abstract ▾
Solar and Stellar Astrophysics
Magnetohydrodynamic coronal modelling with high-fidelity magnetograms
Haopeng Wang, Stefaan Poedts, Wensi Wang et al. (+16 more)
To resolve coronal structures in detail, it is essential to constrain the coronal model using high-fidelity magnetograms with minimal smoothing. In this paper, we improve the numerical stability of COCONUT to enable efficient coronal MHD simulations driven directly by photospheric magnetograms, including
magnetic field strengths exceeding 1000 Gauss (G), and then use it to investigate the impact of small-scale
magnetic field structures that are typically filtered out of photospheric magnetograms on the coronal simulation results. We constrain this model with two types of magnetograms: the high-fidelity HMI photospheric magnetogram with only minor preprocessing through Gaussian smoothing, which preserves
magnetic field strengths reaching 1500 G, and one processed using 50th-order filtered spherical harmonics, where the
magnetic field strength is limited to below 100 G. The results demonstrate that the COCONUT can reproduce the large-scale coronal structures revealed by pB and EUV observations. Compared with the coronal MHD simulation constrained by a conventionally filtered magnetogram, the simulation constrained by the high-fidelity magnetogram yields approximately $50\%$ more OMF and produces EUV images that are more consistent with the observations. Moreover, the results show that when viewed from low latitudes in certain EUV passbands, such as the 193~Å channel, open-field regions near the
solar poles may appear as bright structures that are typically associated with closed-field regions. This finding highlights the need for
solar polar-orbiting missions to overcome the limitations and ambiguities arising from the restricted view of the polar regions from the ecliptic plane and to provide more complete full-Sun observations.
Read the full abstract ▾
All authors Haopeng Wang, Stefaan Poedts, Wensi Wang, Rayan Dhib, Andrea Lani, Martina Condoluci, Tinatin Baratashvili, Quentin Noraz, Luis Linan, Hyun-Jin Jeong, Junyan Liu, Zhenguang Huang, Rui Liu, Lingyu Dong, Rui Zhuo, Liping Yang, Giovanni Alejandro Pinzón Estrada, Jasmina M. Magdalenić Zhukov, Brigitte Schmieder
Machine Learning
Certification-Enhanced Generalization Bounds
Leo Elmecker-Plakolm and Matthew Wicker
We investigate the use of formal methods to provide tight and sound generalization bounds for learning
algorithms. By casting the traditional notion of algorithmic stability as a specification to be verified, we demonstrate that recent advances in reachability analysis can yield provable bounds on the generalization of a given model and
algorithm on a sample
dataset. As sample-specific algorithmic stability is insufficient to bound the usual distributional notion of generalization, we develop a novel concentration inequality to connect the sample-specific results of formal certification
algorithms to the required distributional analysis for bounding the expected generalization gap. The resulting framework enables the analysis of prior generalization bounds to extend far beyond their original restrictive assumptions. Our approach computes sound bounds on the expected generalization gap in a constant number of
algorithm runs without making any analytical assumptions on the
algorithm; to achieve non-vacuous bounds we only require that the certified reachable parameter set is bounded — a condition that we do not assume but formally verify. In practice, we demonstrate that our framework provides formal generalization guarantees that are orders of magnitude tighter than alternative sound computational approaches at scales ranging from toy
datasets to fine-tuning classification heads on top of modern large
language models. While we implement certification-enhanced versions of several well-known stability results, future extensions of our approach will enable tighter bounds and enhanced practical adoption across the spectrum of modern generalization bounds.
Read the full abstract ▾
Statistics Theory
Bradley-Terry model under general comparison graphs
Ruijian Han, Yuanhang Luo and Yiming Xu
The Bradley-Terry model is a parametric model for ranking from pairwise comparisons. Existing asymptotic theory for the maximum likelihood
estimator (MLE) in regimes where the number of objects grows often requires homogeneity assumptions or compatibility conditions on comparison graphs, which limits its applicability to many practical settings. In this work, we establish uniform consistency of the MLE under general deterministic comparison designs. Our pairwise error bound consists of a pair-specific term governed by the effective resistance between the corresponding objects and a global term governed by the spectral gap of the unnormalized graph
Laplacian. The bound is tight up to logarithmic factors for certain graph sequences. For any prespecified sequence of pairs satisfying an additional balancing condition, we further establish asymptotic normality of the corresponding estimated utility differences. As an application, we obtain uniform consistency for independent-edge random graph designs whenever the minimum edge probability exceeds the Erdős-Rényi connectivity threshold by a logarithmic factor, as well as asymptotic normality under additional balancing conditions.
Read the full abstract ▾
Cryptography and Security
DP-ES: Differentially Private Evolution Strategies for Prompt Optimization
Ziniu Liu, Aiping Li, Yue Han et al. (+5 more)
Token-level differentially private (DP)
prompt optimization methods such as DP-OPT can become unstable under tight
privacy budgets: on GSM8K, DP-OPT obtains $49.5\pm28.5\%$ across 30 runs, and a logged search trajectory reveals
prompt-template drift and noise-sensitive irreversible choices. We diagnose these as structural consequences of greedy
token-by-
token construction over privately aggregated counts. We then propose DP-ES (Differentially Private
Evolution Strategies), a structurally cleaner alternative that maintains a population of full
prompts, mutates them via LLM calls that never access the private
dataset, and spends
privacy only on sampled-Gaussian evaluation; deterministic or Gumbel-smoothed selection is post-processing. Under a conservative $(\varepsilon\leq1.0,\delta=10^{-5})$ guarantee, DP-ES achieves 88.1% on GSM8K (+38.6 pp over DP-OPT, approximately 9 times lower standard deviation), 99.7% on MedQA, 73.5% on BANKING77, and 86.8% on Alpaca. It is also 2.5 times faster in wall-clock time and uses 3.3 times fewer logged private-data call groups than DP-OPT. Selection and population ablations, implementation-level noise checks, and a 200-profile exact-match memorization stress test complement the formal guarantee. Scope: Our experiments establish optimization robustness under DP noise, especially where
prompt structure is critical; end-to-end validation on genuinely sensitive, non-saturated deployment data remains future work.
Read the full abstract ▾
All authors Ziniu Liu, Aiping Li, Yue Han, Han Yu, Junjian Zhang, Dong Zhu, Changjian Li, Shiqiang Zhang
Robotics
Encoded but Not in Control: Revealing the Grounding Gap in Vision-Language Robot Policies
Shaohan Jiang, Jiahang Cao, Qiduo He et al. (+9 more)
Instruction following is central to
language-conditioned
robot policies:
language should determine what to do when the same scene permits multiple valid actions. Yet successful execution alone cannot establish whether a policy follows the instruction or infers the task from the scene. We study this ambiguity through scene-preserving instruction interventions, using valid target substitutions, arbitrary nouns, and unrelated sentences while holding the scene fixed. We evaluate
vision-
language-action (VLA) policies and world-action models (WAMs) in simulation and in real-world experiments. Our analysis addresses three questions: (a) Does task success imply instruction following? When instructions request a different visible object, all evaluated policies predominantly approach and pick up the incorrect original target associated with the scene. (b) Is this failure caused by
language insensitivity? Instruction perturbations affect task performance. A layerwise action lens shows intermediate action predictions respond to these perturbations. Linear probes accurately recover instructed targets, indicating modified instructions are encoded despite rarely determining target selection. (c) Why does encoded
language fail to control action?
Attention analysis indicates weak instruction-
token contributions to action generation. Target-
token attention can remain focused on the original object, revealing a mismatch between target encoding and visual grounding. UMAP and shared non-negative
matrix factorization show target information remains accessible within representations increasingly organized by scene identity. Our findings expose a grounding gap concealed by nominal success and provide a diagnostic framework. They further establish a concrete criterion for progress: policies should reliably follow valid changes in user intent, even when they conflict with scene-favored behavior.
Read the full abstract ▾
All authors Shaohan Jiang, Jiahang Cao, Qiduo He, Fengting Deng, Kun Wu, Jingkai Sun, Jiaxu Wang, Qiang Zhang, Qihao Zheng, Chunfeng Song, Ping Luo, Andrew F. Luo
Computer Vision
Joint Class-Time Learning for Video Classification with Multi-Instance Partial-Label Learning
Lingyu Shen, Wei Tang, Fakhri Karray et al. (+1 more)
Multi-instance partial-label learning (MIPL) addresses inexact supervision in both the instance and label spaces, which can be applied to
video classification. However, bag-level labels do not explicitly supervise the correspondence between candidate classes and temporal evidence. We propose {\ours}, which couples label disambiguation with temporal evidence allocation through a joint class–time assignment. Occupancy-regularized spherical matching associates contextualized
video features while learning nonuniform temporal mass and discouraging excessive concentration. During training, candidate-restricted inference recomputes the assignment within the candidate label set. A dual-marginal KL projection then constructs a structured teacher that incorporates momentum-refined class beliefs while preserving the proposal's temporal occupancy. A single plan-level KL objective aligns the full-space predictor with this teacher. Our analysis characterizes when candidate re-solving differs from masking and shows that, under the stated construction, the joint objective decomposes into class-marginal and class-conditional temporal supervision. We construct VCMIPL
benchmarks from Breakfast, DoTA, and FineAction using model-generated candidate labels and evaluate the method across four feature representations. Extensive experimental results demonstrate that PIVOTMIPL outperforms existing MIPL
algorithms in both effectiveness and efficiency.
Read the full abstract ▾
All authors Lingyu Shen, Wei Tang, Fakhri Karray, Min-Ling Zhang
Statistical Methodology
On the nineteenth-century origins of significance testing and p-hacking
Glenn Shafer
Although the names significance test, p-value, and confidence interval came into use only in the 20th century, the methods they name were already used and abused in the 19th century. Knowledge of this earlier history can help us evaluate some of the ideas for improving statistical testing and estimation currently being discussed. This article recounts first the development of statistical testing and estimation after Laplace's discovery of the central limit
theorem and then the subsequent transmission of these ideas into the English-
language culture of mathematical statistics in the early 20th century. I argue that the earlier history casts doubt on the efficacy of many of the competing proposals for improving on significance tests and p-values and for forestalling abuses. Rather than further complicate the way we now teach statistics, we should leave aside most of the 20th-century embellishments and emphasize exploratory data analysis and the idea of testing probabilities by betting against them.
Read the full abstract ▾
Robotics
AUTOPILOT An Advanced Perception, Localization and Path Planning Techniques for Autonomous Vehicles Using YOLOv7 and MiDaS
Harshkumar Devmurari, Gautham Kuckian and Prajjwal Vishwakarma
Self driving vehicles have emerged as a reliable technology that has the capability to transform transportation and mobility. The development of self driving cars requires significant advances in a number of areas, including perception, localization, decision making, and control. This research paper is based on the project implementation of the combination of object detection using YOLO (You Only Look Once), depth sensing using MiDaS for the localization and perception of obstacles, perspective transform, and decision making for path planning in self driving cars. The contemporary state of the technology for object detection, depth sensing, localization, and path planning evaluates the performance of the combined system through simulations and experiments. The results show that the combination of YOLO and MiDaS provides a new robust system for object detection and depth sensing. This research paper contributes to the advancement of self driving car technology and provides new and innovative approaches to the perception and localization of obstacles in the environment. Keywords: YOLO, MiDaS, perception, localization, decision making
Read the full abstract ▾
Quantum Physics
Unitary RQL Equals RQL
Quinten Tupker
Intermediate measurements let
quantum computations discard information and reuse their workspace. Deferring all measurements can require storing their entire history, which need not preserve logarithmic space. Fefferman and Remscrim proved that measurements can nevertheless be eliminated with two-sided bounded error, and asked whether the same holds with one-sided error [FR21]. We prove RQL{\Gamma} = RQUL{\Gamma} for the standard gate set {\Gamma} = {H, T, CNOT}, preserving
polynomial time, logarithmic space, and exactly zero acceptance on no-instances. Our proof extends the density-
matrix doubling method of Girish, Raz and Zhan [ GRZ21 ] to general channels, including resets and classical
memory. We also prove measurement elimination for broader finite gate sets with exact inverses, including suitable gates with transcendental entries, and establish gate-set independence for a specified family of algebraic gate sets. A separate history-checking construction proves QMAL1,G = QUMAL1,G under explicit gate assumptions: measurements can also be eliminated from logarithmic-space
quantum verification while preserving perfect completeness
Read the full abstract ▾
Machine Learning
Parameter Estimation in Machining Dynamics with Regenerative Delay and Nonsmooth Friction using Physics-Informed Neural Networks
Meiyazhagan Jaganathan, Vikram Pakrashi and Aasifa Rounak
A multi-domain eXtended Physics-Informed Neural Network (XPINN) framework is developed for nonsmooth Delay Differential Equations (DDEs). This is the first implementation to demonstrate the efficacy of partitioning the temporal domain into subdomains of integer multiples of the characteristic time delay and progressively training the associated subnetworks while freezing previously learned parameters. The efficacy of the proposed framework is demonstrated using a machining dynamics model that incorporates both regenerative and nonsmooth frictional effects. Results demonstrate that the proposed multi-domain XPINN framework leads to better solution reconstruction in DDEs and improved parameter estimation compared to a generic PINN (SPINN) formulation. The proposed method works particularly well for extended temporal domains and non-constant history functions. The robustness of inverse XPINN (I-XPINN) is also assessed using reference data contaminated with Gaussian measurement noise. Results indicate that I-XPINN remains resilient to measurement noise and the physics-informed constraints guide the network toward accurately recovering the underlying dynamics. This demonstrates, for the first time, the potential of the proposed framework for reliable parameter identification in DDEs characterised by nonsmoothness and large time delays.
Read the full abstract ▾
Machine Learning
SO(3)-RoPE for Spherical Transformers
Christian Libner, Chase van de Geijn, Alexander S. Ecker et al. (+1 more)
Spherical data arise in many scientific applications. Often spherical
transformers disregard the geometry of the underlying spherical domain, causing distortions and coordinate singularities near the poles. We introduce SO(3)-RoPE, a relative positional
embedding that incorporates spherical geometry into
transformer attention through
unitary SO(3) representations. Our formulation is SO(3)-equivariant and compatible with FlashAttention, retaining efficiency of vanilla
transformers. On shallow water dynamics prediction over a rotating sphere, our SO3ViT outperforms an S2
Transformer baseline with lower errors and reduced runtime.
Read the full abstract ▾
All authors Christian Libner, Chase van de Geijn, Alexander S. Ecker, Maurice Weiler
Data Structures and Algorithms
Hiding a Vertex from the Temporal Explorer: A Lower Bound for Degree-Bounded Temporal Graphs
Daniele Carnevale
A temporal graph is a sequence of graphs on a common set of $n$ vertices, its snapshots, one for each time step. An
agent, knowing the entire sequence in advance, may at each time step wait or move along an edge of the current snapshot, and the temporal graph is explored once the
agent has visited every vertex. We consider always-connected temporal graphs, in which every snapshot is connected, and ask how long exploration can be forced to take when the underlying graph, the union of all snapshots, has maximum degree at most~$\Delta$. Two lower bounds were known in this setting, $\Omega(\Delta n)$ and $\Omega(n\log n)$, realised by different constructions. We establish a stronger lower bound, answering a question of Bastide, Groenland, Michel and Rambaud. Specifically, for every $n$ and $\Delta$ with $\Delta_0\le\Delta\le n-1$ we construct an always-connected temporal graph on $n$ vertices with underlying maximum degree at most~$\Delta$ that cannot be explored in fewer than $\gamma\,\Delta n\,(1+\log(n/\Delta))$ time steps from any start vertex, where $\gamma>0$ and $\Delta_0$ are absolute constants. The construction is deterministic, and every snapshot is a spanning tree with exactly one vertex of degree greater than three. Time is divided into phases. In each phase, a rotating-cycle gadget prevents the
agent from reaching more than half of the trees attached to it. Between phases, we reassign target vertices among these trees using walks on a constant-degree expander, and a Gray code order of the targets keeps the underlying degree bounded. The reassignment guarantees that, for any walk of the
agent, some target remains unvisited throughout all phases.
Read the full abstract ▾
Quantitative Methods
APOD: reasoning-guided agentic population ordinary differential equation discovery for pharmacological digital twins
Romain Ferrara, Martin Soucail, Victor Gertner et al. (+7 more)
Establishing ordinary differential equations (ODEs) describing population data is a fundamental part of mathematical modeling in pharmacology, crucial to developing digital twins. However, doing so from sparse, noisy data is a slow, expert-driven task. Existing automated methods either search a restricted model space or ignore population inter-individual variability. Here we introduce APOD (Agentic Population ODE Discovery), a
language-model
agent that iteratively reasons over biological knowledge and fit diagnostics in an open-ended search space to discover a population digital twin (PDT), i.e., a shared ODE system with between-subject variability. On synthetic pharmacokinetic and tumor-dynamics
benchmarks, APOD recovered ground-truth structures in 94-100% of runs, 12-fold faster in median than an established library-based search. On real cohorts it converged to valid structures, and proposed a PDT of radioligand-therapy-induced platelet dynamics that predicts thrombocytopenia from first-cycle data and simulates alternative dosing schedules that lower the predicted risk of toxicity.
Read the full abstract ▾
All authors Romain Ferrara, Martin Soucail, Victor Gertner, Adil Moussali, Joris Cocquebert, Sandrine Oziel-Taieb, Julien Nicolas, Florence Gattacceca, Mihaela van der Schaar, Sébastien Benzekry
Machine Learning
LeAVJEPA: A Minimalist Architecture for Audio-Visual Self-Supervised Learning
Benjamin Robson, Santeri Mentu, Wenshuai Zhao et al. (+1 more)
Prior audio-visual self-supervised learning methods rely on mechanisms such as EMA target encoders, prediction heads, reconstruction decoders, and contrastive losses. We introduce LeAVJEPA, the first audio-visual encoder trained under LeJEPA's collapse-free objective. A single early-fusion
Vision Transformer processes audio,
video, and joint audio-
video inputs. Modality dropout treats a missing modality as another view of the same event, making cross-modal
alignment implicit in the objective. The model aligns global
embeddings with modality-specific local
embeddings, and SIGReg prevents representational collapse. A controlled ablation identifies modality dropout as the key mechanism for audio-visual
alignment. Despite the architectural simplicity, LeAVJEPA reaches 36.0 mAP on AudioSet-20K and 91.3% accuracy on ESC-50 under frozen evaluation. After fine-tuning, it reaches 61.1% accuracy on VGGSound, and its
embeddings support zero-shot audio-visual
retrieval.
Read the full abstract ▾
All authors Benjamin Robson, Santeri Mentu, Wenshuai Zhao, Arno Solin
Algebraic Geometry
Lifting negative curves from finite fields and Seshadri constants at ten points
Piotr Pokora and Tomasz Szemberg
We use finite-field calculations and mixed-characteristic deformation to construct nef divisors on very general complex blow-ups of the plane. A lifting and smoothing criterion converts a reducible nodal divisor over $\mathbb{F}_{127}$ into a smooth genus-$10$ curve of class $117H-37(E_1+\cdots+E_{10})$, retaining control of its first-order obstruction map. Its normal bundle in a suitable deformation is semistable, so Petrakiev's multiplicity estimate yields the nef class $27379H-8658(E_1+\cdots+E_{10})$ at ten very general points. Their multipoint Seshadri constant is therefore at least $8658/27379$, improving Petrakiev's bound $228/721$. The
geometric input is an existence criterion in characteristic zero verified by computations in positive characteristic.
Read the full abstract ▾
High Energy Physics (Theory)
Momentum Flux for Small-Amplitude Oscillons
N. S. Manton and Azadeh Mohammadi
An oscillon in scalar field theory in $(1+1)$-dimensions is an oscillating, localised, almost periodic classical solution that slowly radiates and decays to the
vacuum solution. We calculate its momentum flux integrated over a period, which would be zero for an exact breather. For an oscillon of small amplitude $\eps$, truncated at order $\eps^N$, we show that the integrated momentum flux is of leading order $\eps^{N+4}$ for even $N$ and $\eps^{N+5}$ for odd $N$, except for $N=1$, where it is of order $\eps^4$. It is therefore very small.
Read the full abstract ▾
General Relativity and Quantum Cosmology
Black Hole Horizon Fluctuations from Unruh Quantum Noise
Kashif Ammar Yasir
The mean Hawking flux does not determine how
quantum stress fluctuations perturb a black-hole geometry. That information resides in field correlations, including
vacuum contributions absent from the outgoing luminosity. Here we connect the complete exterior Unruh Wightman function to a finite-band
stochastic reconstruction of Schwarzschild geometry. We retain the incoming
vacuum and both outgoing frequency branches, form the connected stress covariance before angular projection, and propagate its monopole through a density-driven radial response. A shared Gaussian reconstruction then resolves how the value and derivative of the lapse control different
geometric observables. For the specified covariance and noise amplitude, 2,200 realizations give relative widths of 0.0185 for the local horizon radius and 0.0483 for the surface-
gravity temperature parameter, consistently across histograms and reconstruction checks. Covariance derivatives explain this unequal sensitivity. The construction separates the
quantum-state input from the gravitational closure and spatial sampling assumptions, providing a reproducible finite-resolution description of how Hawking-field correlations affect horizon positions and surface-
gravity statistics.
Read the full abstract ▾
Astrophysics of Galaxies
Broadband radio continuum and host galaxy properties of extremely inverted spectrum compact radio galaxies
Mukul Mhaskey, Barenya Dev, Aritra Basu et al. (+3 more)
We present broadband
radio continuum spectra spanning 0.15-8 GHz for a sample of ten Extremely Inverted Spectrum Extragalactic
Radio Sources (EISERS), selected from narrowband, low-frequency surveys below 1.4 GHz. In these surveys, EISERS show unusually steep low-frequency spectral indices, approaching or exceeding the synchrotron self-absorption (SSA) limit of $\alpha$= +2.5, expected for homogeneous synchrotron sources. We have performed follow-up quasi-simultaneous
radio observations of these sources using uGMRT (0.150-1.4 GHz) and VLA (1.0-8.0 GHz). The new observations provide improved frequency coverage and consistent flux calibration across both the optically thick and optically thin spectral regimes. We find that the spectral slopes of EISERS in optically thick regime are flatter than the two frequency spectral indices obtained using TGSS and WENSS and lie in the range +1.6 to +3, suggesting that heterogeneous survey data can introduce biases through flux-scale offsets, intrinsic variability of these compact sources and diverse angular resolutions between surveys. We modelled the broadband
radio spectra as synchrotron emission affected by synchrotron self-absorption and subsequently modified by free-free absorption in an external ionised medium. The estimated physical parameters based on the best-fits indicate that the
magnetic fields and thermal electron densities are similar to the general class of peak spectrum sources, with $B_{SSA}$ spanning over a large range of values from 20 mG to 8 G and $n_{e} \sim 29-128{cm}^{-3}$. Optical and infrared analysis further indicates that the EISERS are preferentially associated with dusty starburst
galaxies, composite starburst-AGN systems, or obscured AGN embedded in dense circumnuclear environments, supporting a scenario in which at least a fraction of these sources represent young or environmentally confined
radio AGN.
Read the full abstract ▾
All authors Mukul Mhaskey, Barenya Dev, Aritra Basu, Helmut Meusinger, Surajit Paul, Matthias Hoeft
Computer Vision
Frequency-Decoupled Diffusion Guidance for Non-Blind Image Deblurring
Sihan Wang, Jinshu Huang, Haibin Su et al. (+1 more)
Pretrained
diffusion models provide powerful image priors for training-free posterior sampling in image restoration. To guide this sampling process, frequency-aware methods progressively incorporate measurement information across frequency bands, facilitating coarse-to-fine reconstruction. However, existing methods typically do not explicitly separate frequency activation from degradation-induced attenuation, leaving attenuation differences among inactive frequencies insufficiently modeled. In this work, we propose frequency-decoupled posterior guidance to separate frequency activation from attenuation-aware spectral regularization. Specifically, a progressive low-to-high frequency schedule determines the active measurement band, while a kernel-derived attenuation map defines a selective spectral prior over inactive components. To stabilize the sampling process, we also introduce a local trajectory regularizer that suppresses spatially irregular state-to-clean deviations. For a fixed endpoint energy, we provide a KL-regularized path-space interpretation. In practice, we construct time-dependent guidance through local energy corrections using a Tweedie plug-in approximation. Experiments on natural-image
benchmarks demonstrate strong PSNR and SSIM performance across challenging non-blind deblurring settings, even at higher measurement noise levels.
Read the full abstract ▾
All authors Sihan Wang, Jinshu Huang, Haibin Su, Yunhua Xue
Cosmology and Nongalactic Astrophysics
Gravitational-wave background from supermassive black holes without merger trees: Tidally regulated mergers and direct inference from the NANOGrav 15 yr data
Shin'ichiro Ando
We present a semi-analytic framework that computes the gravitational-wave background (GWB) from supermassive black hole (SMBH) binaries by direct integration of an $N$-body-recalibrated extended Press-Schechter
accretion kernel, bypassing Monte Carlo merger trees entirely. The calculation is resolution-free and fast enough to be sampled directly against
pulsar-timing data with no emulator or precomputed model bank. Between
halo accretion and black-hole binary formation, we couple the satellite's simulation-calibrated
tidal mass loss continuously to Chandrasekhar dynamical friction. Satellites below a sharp, nearly
redshift-independent mass-ratio threshold, $\xi \simeq 0.04$, are stripped faster than they sink and stall: their drag collapses onto the
stellar core, and 38% of accreted satellites are unmerged by $z=0$. Combining this population model with per-binary
stellar-hardening turnovers and eccentric spectra, we sample the five-parameter posterior directly against the NANOGrav 15 yr Hellings-Downs-correlated free spectrum, excluding the $15.8$ nHz bin, whose excess power a subsequent chromatic-noise reanalysis attributes to
pulsar noise. The data prefer a black-hole mass normalization $+0.31^{+0.20}_{-0.23}$ dex above the locally calibrated black-hole-bulge mass relation, consistent with earlier indications that the nanohertz background is loud relative to local SMBH calibrations and robust to the delay model, merger kernel, local mass calibration, and its
redshift evolution. The spectrum is compatible with, but does not require, eccentric binaries or dense
stellar environments.
Read the full abstract ▾