High Energy Physics (Phenomenology)
Precise QCD Predictions for the Scotogenic Model at Colliders
Mohammed Boukidi, Camila Ramos and Richard Ruiz
Motivated by the anticipated data rates of the high-luminosity phase of the Large
Hadron Collider, we report the first predictions up to next-to-leading order in
quantum chromodynamics with parton-shower matching for the production and decay of scalars and heavy
neutrinos from the Scotogenic model for
neutrino masses and dark matter at $\sqrt{s}=14$ TeV and $\sqrt{s}=100$ TeV. We carry out our simulations using the FeynRules+MadGraph5\_aMC@NLO+Pythia8 pipeline. As a byproduct, we report the public release of the SM\_Scoto Universal FeynRules Object libraries for modeling the Scotogenic model in high-energy processes. We comment on prospects of observing new rare $Z$ decays at future $e^+e^-$
collider facilities as well as on the production of new particles at a multi-TeV $\mu^+\mu^-$
collider.
Read the full abstract ▾
High Energy Physics (Theory)
Collapse to extremality: The third law of black hole thermodynamics in quantum gravity
Gustavo J. Turiaci and Chih-Hung Wu
We present a two-step construction to violate the classical third law of black hole thermodynamics for charged black holes in four dimensions. A near-extremal black hole with an $\mathrm{AdS}_2\times X$ throat is first prepared and then driven to extremality in finite time by a charged shell. This construction suggests that the second step is universal for near-extremal black holes, and can be used to construct violations of the third law for Kerr black holes. The same description allows us to study the process in
quantum gravity. The classical collapse condition selects a BF-violating matter sector, for which we construct a
holographic dictionary and derive the
quantum transition amplitude.
Quantum effects replace the classical extremal endpoint by a probability distribution that vanishes at extremality and peaks at positive energy. This obstruction is consistent with the absence of non-supersymmetric extremal black hole states in
quantum gravity. We show that driving the peak close enough to extremality for
quantum gravity to become important requires a preparation time that scales with the black hole
entropy.
Read the full abstract ▾
Earth and Planetary Astrophysics
Revealing the Shiny Nature of Lava Worlds: Small Planet Phase Curves with TESS
Brandon Park Coy, Michael Zhang, Fei Dai et al. (+1 more)
Recent thermal emission observations suggest that `lava worlds'–Earth-sized
exoplanets hot enough to melt their silicate surfaces–have atmospheres on a population-level. This evidence comes from measured dayside brightness temperatures much lower than expected for a dark, maximally hot bare rock. However, these results are degenerate between a thick atmosphere advecting heat to the
planet's nightside or a high albedo reflecting incoming radiation. This degeneracy can be resolved through visible light phase curve observations, which can constrain how much light is reflected by the
planet. Here, we present a population-level analysis of the visible-light TESS phase curves of $R\lesssim5\,R_{\oplus}$
exoplanets. We select the population by defining a `Reflection Spectroscopy Metric' meant to quantify the relative signal-to-noise of reflected light observations. We report new $>3\,\sigma$ detections of the TESS phase curve signals of LTT 9779 b and the lava worlds TOI-2431 b, TOI-561 b, 55 Cancri e, K2-141 b, and TOI-1444 b. We use these results and previous results from Kepler and CHEOPS to homogeneously derive
geometric albedos for these
planets, revealing that lava worlds generally show high albedos compared to the low
geometric albedos measured for hot Jupiters at similar temperatures. These results support an emerging picture of reflective silicate clouds on these worlds cooling their daysides. The upcoming PLATO mission is expected to tightly constrain the reflected light phase curves of lava worlds.
Read the full abstract ▾
All authors Brandon Park Coy, Michael Zhang, Fei Dai, Qiao Xue
Optimization and Control
Silver Rate Is (Almost) Optimal for Gradient Descent: The Strongly Convex Case
Kaizhao Liu and Yuhan Ye
We study gradient descent with predetermined nonnegative stepsizes on smooth strongly convex functions. Let $p_{\mathrm{sil}}=\log_2(1+\sqrt2)$ and $\kappa$ be the condition number. We prove the iteration lower bound $\Omega\left(\kappa^{\frac{1}{p_{\mathrm{sil}}}-o(1)}\log\frac1\delta\right)$ for both relative squared distance and relative function error, uniformly over $0<\delta<1$ and sufficiently large $\kappa$. This matches the
polynomial exponent of $\kappa$ for the Silver stepsize schedule established in [Altschuler and Parrilo, 2025].
Read the full abstract ▾
Robotics
MSK-Bench: Benchmarking Full-Body Musculoskeletal Motor Control Across Tasks, Control Paradigms, and Physiological Metrics
Mengtao Ou, Zongzheng Zhang, Zhenghao Xiao et al. (+7 more)
Musculoskeletal (MSK)
humanoids provide a physiologically grounded embodiment for studying full-body motor control, but their high-dimensional muscle actuation, delayed activation dynamics, and redundant muscle–tendon structures make learning substantially harder than torque-driven
humanoid control. Existing MSK
benchmarks remain fragmented across gait, prosthetics, dexterous hands, or challenge-specific tracks, leaving full-body muscle-actuated control insufficiently evaluated under standardized tasks, methods, and metrics. We introduce MSK-Bench, a
benchmark of 22 full-body motor-control tasks organized into three progressively challenging categories: postural stabilization, common locomotor behaviors, and contact-rich environmental interaction. Under unified task protocols and robustness perturbations, MSK-Bench evaluates 5 representative control paradigms, including reward-based RL, agentic reward tuning, latent-action RL, imitation-prior control, and residual adaptation over imitation priors. Beyond task success and reward, MSK-Bench further reports robustness analysis and physiology-oriented diagnostics, including activation cost, joint smoothness, and EMG-envelope similarity. Our empirical study shows that embodiment-aware exploration and structured action representations improve task coverage in high-dimensional muscle spaces, imitation priors enhance reference-compatible stabilization and locomotion but degrade under contact-rich terrain mismatch, and residual adaptation can recover successful behaviors when fixed references fail. We further find that improved task success does not necessarily imply improved physiological agreement, highlighting the importance of evaluating task performance, robustness, and physiological behavior jointly. MSK-Bench provides a task–method–metric testbed for full-body muscle-actuated
humanoid control.
Read the full abstract ▾
All authors Mengtao Ou, Zongzheng Zhang, Zhenghao Xiao, Yixuan Pan, Ziwen Zhuang, Hang Zhao, Hongyang Li, Yanan Sui, Libin Liu, Hao Zhao
Soft Condensed Matter
Experimental Characterization of the Passive Mechanical Behavior of Rat Pelvic Floor Muscles: Effects of Test Conditions and Pregnancy
Jose L. Monclova, Matthew Shankel, Ava M. English et al. (+2 more)
Purpose: Pelvic floor muscles undergo large deformations during vaginal delivery, yet the passive tensile behavior of pregnant and non-pregnant pelvic floor muscle and the influence of testing protocols remain incompletely characterized. This study evaluated how pregnancy, loading rate, preconditioning, and storage affect passive tensile mechanics of rat pelvic floor muscle. Methods: Levator ani samples from pregnant and non-pregnant rats were tested in uniaxial tension to failure across multiple preconditioning amplitudes, two loading rates, and fresh or frozen storage conditions. Peak and transition metrics, strain energy density, toe-region stiffness, and high-strain stiffness were extracted using automated region-detection methods. Rat properties were also compared with digitized human cadaver levator ani data. Results: Pregnant rat samples reached greater peak strains than controls, although the difference was not significant. Preconditioning affected peak strain in a loading-rate-dependent manner: responses tapered near 2% preconditioning strain at 0.5% strain/s, whereas no comparable tapering occurred through 15% at 1% strain/s. Compared with human cadaver data, rat samples reached substantially larger peak strains and lower high-strain stiffness. Conclusion: This study provides one of the first direct experimental comparisons of passive pelvic floor muscle mechanics between pregnant and non-pregnant tissue, while systematically evaluating how multiple testing-protocol variables influence the measured response. Distinguishing protocol-dependent variability from biological differences provides a more reproducible basis for cross-study comparison and constitutive model calibration. Rat pelvic floor muscle is a useful model for pregnancy-associated adaptation, but absolute mechanical values should be translated to human tissue cautiously.
Read the full abstract ▾
All authors Jose L. Monclova, Matthew Shankel, Ava M. English, Mona Eskandari, Heidi P. Feigenbaum
High Energy Physics (Phenomenology)
The LHC is not enough: the LZ High-Recoil Event at FCC-hh and a Muon Collider
Benedikt Maier and Michael Spannowsky
The high-energy
nuclear-
recoil candidate reported by LUX-ZEPLIN could point to dark matter beyond the reach of the LHC. We examine how future
colliders could test this interpretation, starting with the thermal Higgsino and extending to a broad class of electroweak multiplets containing neutral and charged particles. We assume an approximately pure multiplet protected by an exact stabilizing
symmetry, with dominant inelastic Z exchange and the full dark-matter abundance set by gauge-dominated thermal freeze-out, where mixing and additional interactions affecting production or freeze-out are negligible. In this setting, heavy thermal masses, rare production and invisible neutral transitions make the Higgsino's main
collider obstacles generic across the class, independently of supersymmetry. Charged lifetimes vary, but the
benchmark spectra studied, including radiative doublet splittings, remain beyond the projected HL-LHC reach. We show how FCC-hh and a high-energy
muon collider can overcome these obstacles. Our FCC-hh estimates show sensitivity to the doublets and several larger multiplets through disappearing charged tracks and complementary channels that do not depend on the charged lifetime.
Muon-
collider projections cover the fermionic doublet and triplet at 10 TeV and the full fermionic family considered here at 30 TeV, subject to the stated spectrum and systematic assumptions. This connects a possible underground signal to a concrete program for testing thermal electroweak dark matter at future
colliders.
Read the full abstract ▾
Strongly Correlated Electrons
Easy-plane anisotropy of magnetic fluctuations in UTe2
Zehao Wang, Ewan Scott, David W. Tam et al. (+9 more)
The large directional anisotropy of superconducting UTe2 along the
lattice a, b, c directions in
magnetic susceptibility, upper critical fields, and Knight shift suggests
spin-triplet Cooper pairing. If
spin fluctuations bind the Cooper pair, one needs to determine their anisotropy in
spin space to characterize the superconducting state. Here we use polarised and unpolarised inelastic
neutron scattering (INS) experiments to show that
antiferromagnetic (AFM)
spin fluctuations at the Y point are highly isotropic within the ac-plane but have no observable magnitude along the b-axis. Our results rule out the assumed Ising anisotropy and constrain the possible pairing channels. AFM exchange is no longer pair forming in the triplet channel and the d vector along b is strongly favoured. We capture the experimental observations within a microscopic framework that encompasses the dual nature of 5f electrons. We include itinerant and localized 5f-moments in a tight-binding model fitted to
quantum oscillations data. In congruence with INS, the exchange between localized moments generically consists of a strong FM intra-dimer interaction, maxima close to the Y and T points, mediated by 5p_y and 6d_{3z^2-r^2} orbitals respectively, and magnetically decoupled ab-planes. We find that anisotropy and dimerization of the moments can lead to gapped excitations that, in combination with Landau damping, can produce the observed overdamped response at the Y point. We suggest that enhancement of the response and its observed near-O(2)-
symmetry could both be driven by proximity to a
quantum critical point where the dimer excitation gap closes and gives place to incommensurate helical order. Analysis of the linearized gap equation for a nearest-neighbour pairing interaction favours the B1u triplet with a dominant d-vector component along b, although the possibility of
spin singlet pairing cannot be ruled out.
Read the full abstract ▾
All authors Zehao Wang, Ewan Scott, David W. Tam, Philippe Bourges, Keke Feng, Tyler W. Wannamaker, Paul Steffens, Arno Hiess, Ryan E. Baumbach, M. Brian Maple, Michal P. Kwasigroch, Pengcheng Dai
Robotics
Backdoors in Learning-Based Industrial Robotic Arm Manipulation: An Empirical Security Study
Zijian Zhang, Zhen Zeng, Zhongshu Gu et al. (+1 more)
Learning-based models (e.g., visuomotor and
Vision-
Language-Action (VLA)) are increasingly explored for industrial robotic manipulation, where model predictions are directly translated into physical actions. This tight coupling between model behavior and physical execution makes hidden security vulnerabilities particularly consequential. While
backdoor attacks have been widely studied in conventional AI models, their effects on deployed learning-based robotic arm manipulation systems remain less understood: a backdoored
robot can behave normally during benign operation while inducing attacker-specified behaviors only when specific triggers are present, posing potentially serious risks in physical environments. In this work, we present a preliminary empirical security study of
backdoor attacks and defenses in learning-based robotic manipulation on two real commercial industrial robotic arms (FANUC and xArm). We investigate whether a
backdoor can reliably induce semantically incorrect manipulation behaviors while remaining stealthy under nominal task execution. We then develop an online defense pipeline that detects and neutralizes triggers at runtime, and compare its effectiveness against an offline fine-tuning defense. Beyond defense effectiveness, we further evaluate the computational latency and execution overhead introduced by the defense pipeline to assess its suitability for high-throughput industrial operation.
Read the full abstract ▾
All authors Zijian Zhang, Zhen Zeng, Zhongshu Gu, Sandeep Pisharody
Statistical Methodology
Additive Nonparametric Regression with Spatial and Network Objects
Rajarshi Guhaniyogi, Pritam Dey, Krishnendu Chandra et al. (+2 more)
This article is motivated by an imaging application from the Adolescent
Brain Cognitive Development (ABCD) study, aiming to predict task-based
brain activation maps from t-fMRI using cortical metrics from structural MRI (s-MRI) and
brain connectivity data from resting-state fMRI (rs-fMRI). Hierarchical
Bayesian modeling is well-suited for integrating diverse imaging data and quantifying prediction uncertainty. However, progress in this field is limited due to challenges in designing joint priors that capture the structures and interconnections between different imaging modalities, along with computational complexity and lack of theoretical assurances. To address these challenges, the article introduces a novel regression framework that treats t-fMRI and s-MRI images as functional data, incorporating additive non-linear effects of both network and functional predictors on the functional response. Specifically, we employ Gaussian process (GP) priors on coefficients related to the functional predictors to capture their intricate functional dependencies with the response. Furthermore, a GP prior is assigned to encapsulate the non-linear nodal effects of the network predictor on the response function. The method is supported by theoretical results on predictive accuracy for the functional response, and is empirically validated through simulation studies and analysis of multi-modal neuroimaging data from the ABCD study.
Read the full abstract ▾
All authors Rajarshi Guhaniyogi, Pritam Dey, Krishnendu Chandra, Aaron Scheffler, Bani K. Mallick
Machine Learning
Marginally Correct Tool Caches Can Reverse Group-Normalized Policy Updates
Shivam Gupta
Tool-result caching reduces repeated execution in
agent training, but also couples rollout randomness. We study a two-action model in which independent and shared execution preserve every rollout's conditional reward distribution. Despite this marginal agreement, sharing one
stochastic result per group can reverse the expected group-normalized policy update. We derive an exact finite-group expression: against a constant alternative, the shared update follows the probability of winning minus the probability of losing, rather than the difference in expected reward. A Bernoulli specialization yields a wrong-direction region and a non-vanishing update-variance floor as group size grows. Centering without group standard-deviation scaling preserves the expected-return direction in this model, using an existing
estimator control. Exhaustive finite sums verify 540 configurations and 3,240
estimator evaluations, with a separate ordered-sequence checker. An implementation audit reproduces the sharing path in a pinned, unmodified TVCache stack using 256 scripted rollouts. These results do not measure
language-model training performance or refute TVCache's deterministic-output contract. They establish that marginal output validity alone cannot certify a
stochastic cache as training-equivalent.
Read the full abstract ▾
Human-Computer Interaction
Safety Nudges: User-Facing Interventions for Real-Time AI Risk Awareness
Varshini Elangovan, James Wedgwood, Chhavi Yadav et al. (+3 more)
Conversational AI systems can pose safety risks to their users such as
hallucination, sycophancy, overconfidence, and anthropomorphism, but these risks are difficult for users to detect during everyday use. We introduce Safety Nudges, a browser-based tool that provides lightweight, in situ flags when concerning behavior is detected in chatbot conversations. We evaluated Safety Nudges in a two-week field study with 45 frequent chatbot users, collecting interaction logs, surveys, and feedback on individual nudges. Participants found the tool useful, clear, and minimally disruptive, with nearly all users reporting an increased awareness of potential AI harms, though we found that this improved awareness alone did not necessarily lead to discernible behavioral changes. Our results suggest that user facing safety nudges can complement model-level safeguards by helping people critically evaluate AI responses in context, while highlighting the importance of relevance, calibration, and user control in nudge design for conversational AI safety. The code for our Safety Nudges extension is publicly available at https://github.com/jtbwedgwood/safety-nudges.
Read the full abstract ▾
All authors Varshini Elangovan, James Wedgwood, Chhavi Yadav, William Agnew, Sauvik Das, Virginia Smith
Physics and Society
Network-based modeling of cocaine trafficking flows and displacement effects
H.A. Peters, Frederike Oetker, Robby Roks et al. (+3 more)
The worldwide cocaine
market is undergoing an extraordinary surge. Insights on how cocaine is trafficked from production countries to consumer
markets are limited, and often based on observed trafficking routes (e.g., seizures) alone. In this paper, we introduce a network-based model of actual transnational cocaine flows, beyond flows seized. In this model, cocaine is routed through an international transportation network of land and sea connections, where each link is assigned an interception risk metric for containerized transport. This interception risk metric combines features related to law enforcement inspections, containerized trade, and transport connectivity into three key driving components of interception risk using Principal Component Analysis (PCA). The interception risk metric on each link is then constructed from these driving components, where the importance of each component is characterized by a strength parameter. Given a set of strength parameters, cocaine is assumed to be routed from production countries to consumption countries via routes that minimize the total risk of interception, subject to country-specific supply and demand constraints. In the end, the strength parameters are calibrated by comparing the corresponding flow outcomes with known trafficking activity. From these final trafficking flows associated with the calibrated parameters, we are able to identify `blind spots' (e.g., nonzero flows on links with small or none observed trafficking activity) and provide more evidence of the waterbed effect (e.g., changing routes as a result of increased interdiction efforts by law enforcement).
Read the full abstract ▾
All authors H.A. Peters, Frederike Oetker, Robby Roks, Roy Lindelauf, Robbert Fokkink, Huijuan Wang
Multiagent Systems
Towards Strategy-Level RSI for Skill-Augmented Agents: Learning When to Reuse Skills from Execution Feedback
Shuang Guo, Xiaobo Sui, Fei Jiang et al. (+4 more)
Long-running
agents accumulate reusable Skills, but a Skill that is semantically relevant to a task is not necessarily worth loading in the current state. We study the applicability question that arises once a candidate Skill is known: should it be loaded in the current state? We propose SkillApt, which uses matched WITH/WITHOUT Skill executions on the same task state as persistent evidence, estimates the conditional marginal utility of the Skill, and chooses LOAD or ABSTAIN accordingly. The base model,
agent architecture, and Skill contents stay fixed; only the external deployment policy changes. We call this constrained setting strategy-level recursive self-improvement (Strategy-Level RSI). On 20 Skills and 160 held-out states, as paired evidence accumulates, SkillApt's task success rises from 81.9% under a cold start to 91.3%, matching a strong zero-shot LLM controller; yet SkillApt activates Skills on only 26.3% of states, versus 98.8% for the zero-shot controller. A hard-candidate study shows that non-optimal Skills mostly leave correctness unchanged while raising execution cost, and occasionally cause correctness harm. An ablation shows that a history recording only WITH success makes the policy load almost everywhere, whereas paired evidence substantially improves selectivity. These results indicate that relevance is not applicability: the main effect of execution evidence is not to make the model stronger but to change how existing Skills are deployed, moving the system from near-always loading to selective reuse.
Read the full abstract ▾
All authors Shuang Guo, Xiaobo Sui, Fei Jiang, Ruixin Li, Manxiang Li, Rong Li, Ziyang Ou
Algebraic Geometry
A smooth quintic surface with 85 lines and Picard number 43
Xun Yu and Zigang Zhu
We construct a pencil of complex quintic surfaces with every smooth member containing exactly $85$ lines. We show that one smooth member has Picard number $43$ and by introducing the notion of a $d$-saturated
lattice, we prove that its Néron–Severi group is generated by the lines and four conics. We also construct smooth quintic surfaces with $79$ lines, including examples with $24$ pairwise skew lines, and a smooth surface of degree $10$ with $356$ lines, including $112$ pairwise skew lines.
Read the full abstract ▾
General Economics
Rule-Based Pricing Algorithms and Market Outcomes: An Experimental Study
Adrian Hillenbrand, Hans-Theo Normann, Matthias Potarca et al. (+1 more)
Rule-based pricing tools are widespread in digital commerce, yet we know little about how their design shapes
market outcomes. In a controlled
market experiment, participants use dashboards to build pricing
algorithms competing in a sequential Bertrand game over multiple periods. We vary design features commonly found in commercial repricing tools: warnings about price wars, pre-configured strategies, and advice from a large
language model. Most treatment variations raise
market prices with effects driven by an increase in starting prices and more cooperative
algorithm designs. The results matter for competition policy, platform regulation and current discussions on regulating
algorithm design tools.
Read the full abstract ▾
All authors Adrian Hillenbrand, Hans-Theo Normann, Matthias Potarca, Tobias Werner
Cryptography and Security
Comparative Evaluation of Static Embedding Models for HTTP Request Anomaly Detection
Amanda Riverol, Gustavo Betarte, Rodrigo Martínez et al. (+1 more)
Web applications are increasingly targeted by cyberattacks that exploit HTTP requests to evade security mechanisms. Traditional web application firewalls (WAFs) rely on rule-based approaches that often exhibit high false positive rates and limited adaptability. Recent studies have explored machine learning techniques and word
embedding models to improve anomaly detection in HTTP traffic. This paper presents a
benchmark for static
embedding models, specifically Word2Vec, FastText, and Doc2Vec, within a unified, single-class classification framework. We propose HEDA (HTTP
Embedding-Based Detection Architecture), a modular detection pipeline that combines static
embedding representations with single-class anomaly detection models to detect anomalies at the request level. The approach operates in an unsupervised environment, where both the
embedding models and detectors are trained exclusively on benign HTTP traffic. The proposed methodology is evaluated on three
datasets with heterogeneous characteristics, including both synthetic and real traffic. The experimental results show that the choice of
embedding representation significantly affects detection performance, and that FastText-based embeds produce the most consistent results across all
datasets, achieving high detection rates while keeping false positive rates under control.
Read the full abstract ▾
All authors Amanda Riverol, Gustavo Betarte, Rodrigo Martínez, Álvaro Pardo
Atomic Physics
Quantum Chemistry in a Novel Hybrid Dipolar Atom-Ion Mixture
Claudia Galantini, Mateo Londoño, Luc Verwaal et al. (+3 more)
Merging trapped ions with cold atomic clouds offers intriguing prospects for
quantum chemistry and many-body
quantum simulations. Especially when going beyond the standard alkali atomic baths by using lanthanide atoms, opportunities arise to study the interplay between the intermediate-range atom-ion interaction and the tunable long-range dipolar atom-atom interactions. However, the high total angular momentum of the ground-state of open-shell lanthanides, e.g. $^{5}I_{8}$ for dysprosium, affects the atom-ion potential. Here, we discuss the implications for long- and short-range atom-ion interactions and present a novel apparatus which combines an ytterbium ion (Yb$^{+}$) with dipolar dysprosium (Dy) atoms. We highlight the consequences of this novel Dy-Yb$^{+}$ mixture for observing buffer gas cooling, (non-) radiative charge transfer, and three-body recombination. While the energy-averaged rates are dominated by radiative charge transfer, particularly radiative association, we find that three-body recombination can compete with molecular-ion formation at Dy densities $n\gtrsim 10^{12}$ cm$^{-3}$. This competition is further enhanced by the expected non-thermal distribution of ion energies. These processes could be experimentally characterised through controlled variation of the atom-ion interaction parameters, providing a direct test of our theoretical predictions.
Read the full abstract ▾
All authors Claudia Galantini, Mateo Londoño, Luc Verwaal, Edgar J. D. Vredenbregt, Jesús Pérez-Ríos, Rianne S. Lous
Atomic Physics
A Cryogenic Penning Trap Based on Permanent Magnets
F. Völksen, K. K. Anjum, R. P. Czampiel et al. (+30 more)
We report on a Penning trap based on NdFeB permanent magnets operated at 4\,K with a
magnetic field strength of 280\,mT. Reliable loading and confinement of
protons, H$_2^+$ ions, and electrons was demonstrated with non-destructive single-particle detection sensitivity of
protons and H$_2^+$ ions using superconducting image-current
circuits. The axial frequency of individual particles reaches a shot-to-shot stability of 47 parts-per-billion comparable to that of state-of-the-art precision Penning-trap experiments. Measurements of the
proton modified-cyclotron frequency show a shot-to-shot scatter of $0.14$ parts per million (p.p.m.), presently limited by millikelvin-level temperature fluctuations of the permanent-magnet assembly. We outline a route towards improving this performance by more than an order of magnitude. This development offers broad potential for axial-mode-related precision measurements and cost-efficient Penning-trap experiments, and represents an important step towards compact, scalable, and transportable antiproton-trap systems.
Read the full abstract ▾
All authors F. Völksen, K. K. Anjum, R. P. Czampiel, N. Daehnhardt, S. Gavranovic, J. Hoentges, L. Kürschner, J. D. Regier, D. Schweitzer, 1, 2, P. Schöps, A. M. Thomas, S. Wienhues, B. Arndt, T. Imamura, S. Endoh, B. M. Latacz, M. Leonhardt, P. Micke, J. Morgner, D. Natakala, S. Stahl, H. Yildiz, K. Blaum, J. A. Devlin, Y. Matsuda, C. Ospelkaus, W. Quint, A. Soter, J. Walz, C. Smorra, S. Ulmer
Robotics
FLINT: Fast Lightweight Inference for Traversability
William Bonilla, Maxime Boisvert, David-Alexandre Poissant et al. (+2 more)
Navigation in off-road conditions is challenging due to the lack of structure. There is no fixed vocabulary for what is traversable. The traversability depends on both the environment and the embodiment's dynamics. Neither of these two variables can be hand-labeled at scale. Thus, traversability has to be learned by the embodiment's own experience. Modern platforms tend to use multiple sensors to estimate traversability and navigate: RGBD cameras,
lidar, radar, IMU, with computationally intensive platforms to run inference on neural networks. Against this trend, we propose FLINT, a lightweight traversability
estimator: a 21.6M-parameter backbone, 38\times smaller than a comparable foundation-model backbone, that scores higher on held-out terrain probes and runs at 14.7 FPS on CPU alone using a RGB camera has the only sensor. Despite that gap in scale, FLINT produces a cheaper, more accurate costmap than a deployed foundation-model system (WildOS) on 23 of 24 replayed field logs. We compare different self-supervised learning signals and deploy the resulting models on a real platform in closed-loop field trials: the best self-supervised head reaches 99% autonomy over the route, outperforming a human-label-trained baseline deployed live on the same course. Our results show that heavy sensing and computing are not necessary for traversability estimation.
Read the full abstract ▾
All authors William Bonilla, Maxime Boisvert, David-Alexandre Poissant, David Meger, Louis Petit
Information Retrieval
ItColBERT: An Italian-Specialised Late-Interaction Retriever
Enrico Nello
Neural information
retrieval for Italian is served almost entirely by
multilingual models. Several multi-vector (late-interaction) retrievers include Italian among dozens of
languages, and several strong Italian dense embedders exist, but as of August 2026 no late-interaction retriever specialised on Italian had been released. We present ItColBERT, a 135M-parameter Italian ColBERT trained with PyLate following the ColBERT-Zero recipe: initialise from a checkpoint that already retrieves, then apply supervised contrastive training followed by single-teacher
distillation, for a total of roughly 14.5 GPU-hours on one RTX 3090. Across four Italian
retrieval benchmarks it outperforms every general-purpose late-interaction baseline we tested except one (mLateOn), at 2-4.4x fewer parameters than every baseline but one of comparable size. Our principal empirical finding is methodological and partly negative. On the only cleanly out-of-domain
benchmark (MLDR-it), an inference-time chunking recipe applied to an unchanged checkpoint yields +0.0602 nDCG@10 (p = 0.0225), a larger effect than anything two further rounds of training produced. Self-mined hard negatives and native 1024-
token training were both evaluated against pre-registered decision gates and both failed. We report every comparison with paired bootstrap tests against an empirically measured noise floor of 0.0030 nDCG@10, and we release the weights, the training and evaluation code, and the complete experimental record including the rejected rounds.
Read the full abstract ▾
Machine Learning
QUARTET: Quad-branch cross-Attention and Random-walk Traces for Enhancing Transformers on Relational Graphs
Kyaw Hpone Myint, Nan Jiang, Xiang Li et al. (+4 more)
Relational Deep Learning (RDL) models multi-table databases as heterogeneous temporal graphs, and graph
transformers currently achieve state-of-the-art performance on
benchmarks like RelBench. However, the current leading model, RelGT, suffers from two key limitations: its random local sampler yields loosely connected subgraphs that hinder message passing, and its global
attention module relies on a single, seed-feature-based
memory that ignores broader macro-level dynamics. To overcome these limitations, we introduce QUARTET, an expressive graph
transformer architecture that applies full self-
attention on local subgraphs while enriching global context through cross-
attention branches. Specifically, QUARTET employs a Causal Random Walk (CRW) sampler based on recency-truncated Personalized PageRank (PPR) to extract compact, hub-robust, and densely connected local subgraphs without temporal leakage. Concurrently, a quad-branch cross-
attention module integrates global context from four complementary perspectives: seed feature, seed
topology, temporal dynamics, and collaborative dynamics. Across the RelBench v1 classification tasks, QUARTET consistently matches or outperforms the current state-of-the-art graph
transformer baselines (HGT and RelGT). Ablation studies confirm that the CRW sampler significantly enriches local neighborhood quality, while the global branches provide essential, task-specific predictive gains.
Read the full abstract ▾
All authors Kyaw Hpone Myint, Nan Jiang, Xiang Li, Zhe Wu, Alexandre G.R. Day, Pranab Mohanty, Giri Iyengar
Graphics
SsgCaps: A controlled dataset for the evaluation of sound scene generation algorithms
Modan Tailleur, Junwon Lee, Laurie M Heller et al. (+5 more)
Sound Scene Generation is about the automatic synthesis of artificial sound scenes. We introduce SsgCaps, a publicly available
dataset of human-engineered sound scenes wherein each scene matches a precisely structured
prompt that guides the sampling process. The corresponding
prompts are sampled from a predefined action-based typology that allows extensive sampling while retaining plausibility. SsgCaps is a sound scene
dataset derived from the unpublished reference
dataset for Task 7 of the 2024 DCASE Challenge edition, which contained private-and public-domain audio samples. In contrast, SsgCaps contains only public-domain audio samples, allowing us to open this
dataset to the community. To make this
dataset useful to the community, we first elaborate on the rationale for the
prompt and
dataset structure. We then perform a comparative quantitative analysis of the 2 versions of the
dataset. To do so, we compare both versions to the audio synthesized by the SSG
algorithms submitted to the challenge using Fréchet Audio Distance (FAD) and Kernel Audio Distance (KAD) as well as perceptual ratings. This analysis shows only small differences, which enables us to recommend the open version for further benchmarking of SSG
algorithms.
Read the full abstract ▾
All authors Modan Tailleur, Junwon Lee, Laurie M Heller, Mathieu Lagrange, Keunwoo Choi, Brian McFee, Keisuke Imoto, Yuki Okamoto
Machine Learning
COPE: Continual Personalization of LLMs under Sparse User Feedback via User Embeddings and Self-Evaluation
Ruike Cao, Fugen Yao, Liang Dong et al. (+3 more)
While Large
Language Models (LLMs) have achieved remarkable results across various
benchmarks, their
alignment with normative values often results in homogenized responses that fail to address diverse user preferences. Existing training-free methods often occupy valuable context windows through
prompt engineering, while training-based methods typically remain static post-training, failing to support the continual optimization required in real-world settings. To address these challenges, we propose COPE (Continual Optimization with Personalized
embedding and self-Evaluation), a novel optimization framework tailored for real-world-motivated interaction settings with sparse user feedback. Our framework assigns learnable personalized
embeddings to each user and synergistically integrates preference capture, self-evaluation calibration, and personalized response optimization within a single update step. A key innovation of our method is the use of self-evaluation to generate proxy rewards, enabling continuous model updates even when explicit user feedback is unavailable. Experiments show that COPE consistently outperforms strong training-free and training-based baselines under sparse feedback, and remains complementary to
Retrieval-Augmented Prompting (RAP). Further analyses confirm COPE's reliable self-evaluation, meaningful preference patterns, stable general capabilities, and robustness under shifting preferences and alternative evaluators.
Read the full abstract ▾
All authors Ruike Cao, Fugen Yao, Liang Dong, Jian Xu, Guanjun Jiang, Li Xiao
Differential Geometry
The Second Gap for Minimal Hypersurfaces in the Unit Sphere with Constant Cubic Trace
Shanlin Guan
Let $M^n$, $n\ge3$, be a closed oriented minimal hypersurface in the unit sphere with constant scalar
curvature and constant $f_3=\tr(h^3)$, where $h$ is the second fundamental form. We prove that $S=\norm h^2>n$ implies $S\ge2n$. Equality holds if and only if each connected component is locally congruent to a Cartan minimal isoparametric hypersurface with three distinct principal
curvatures.
Read the full abstract ▾
Discrete Mathematics
A New Method that can Generate Ramsey Colourings for Eight and Thirteen Colours
Charles Gretton, Tomasz Kowalski, Richard Pennifold et al. (+1 more)
We study representations for relation
algebras corresponding to certain edge colourings of complete graphs. Previously suitable colourings were obtained for the number of colours $n$ up to $2000$, with two exceptions: $n = 8$ and $n = 13$. Using a method that we have called the FBF (Fusion Beautiful Fusions) method, we find colourings for $8$ and $13$ colours. Our method is a new guess-and-check approach that we describe as an adaptation of Comer's finite-field method. Using FBF we construct novel colourings that are non-isomorphic to existing published colourings for 5, 6, 7, 9, 10, 11 and 12 colours.
Read the full abstract ▾
All authors Charles Gretton, Tomasz Kowalski, Richard Pennifold, Cody Christopher
Complex Variables
A bootstrap proof of Bohr's theorem
Ole Fredrik Brevig
Let $f$ be an analytic function in the unit disc whose modulus is bounded by $1$. We present a new proof of Bohr's
theorem, which asserts that its majorant function $Mf(r)$ is bounded by $1$ for $0\leq r \leq 1/3$ and that $1/3$ is optimal. The argument also yields a sharp upper bound for $Mf(r)$ in terms of $\lvert f(0) \rvert$, complementing a result due to Bombieri.
Read the full abstract ▾
Solar and Stellar Astrophysics
Spectroscopic Study of the Circularly Polarised Radio Source TYC 2834-1385-1: An Active Binary
Tamara A. S. Bleeker, Alexander Hoogerbrug, Elise Koo et al. (+2 more)
Close binaries with active chromospheres have recently been shown to emit circularly polarised
radio emission at low frequencies, motivating detailed studies of their orbital properties. We present high-resolution ($R=85,000$) optical spectroscopy of TYC 2834-1385-1, a previously unclassified
radio source, obtained with the HERMES spectrograph on the Mercator
Telescope. We identify the source as a double-lined spectroscopic binary and measure radial velocities from both metallic absorption and Ca II emission lines. We find an orbital period of $P=3.601\pm0.002$ days and nearly equal radial-velocity semi-amplitudes for the two components. The system exhibits strong Ca II H and K emission, which is consistent with a BY Dra-type binary containing two similar late-G/early-K main-sequence
stars, with a preliminary classification of K0V. This represents one of the first spectroscopic confirmations of a chromospherically active binary following its identification in a low-frequency
radio survey.
Read the full abstract ▾
All authors Tamara A. S. Bleeker, Alexander Hoogerbrug, Elise Koo, Rudy Wijnands, Joseph R. Callingham
Machine Learning
A Leakage-Aware Multimodal Evaluation Framework for Early Intraoperative Acute Kidney Injury Prediction
Quang Minh Nguyen, Duc Minh Le, Ho Nhat Minh Nguyen et al. (+2 more)
Postoperative acute kidney injury (AKI) after major non-cardiac surgery carries substantial morbidity, yet early intraoperative risk stratification remains difficult. In this retrospective cohort study, we propose SynerT, a waveform-only hybrid temporal backbone that combines a causal dilated TCN with a hierarchy of dilated recurrent layers to encode early intraoperative physiologic trajectories for AKI risk prediction. Building on SynerT, we further design two model variants that extend the backbone with structured clinical context: SynerT-MM, a late-fusion
multimodal extension that integrates hemodynamic burden summaries and preoperative
covariates, and SynerTStack, a leakage-safe stacked ensemble that combines cross-validated predictions from SynerT-MM with strong tabular baselines at the meta-learning stage. All models are evaluated under a strict leakage-aware framework on VitalDB, a high-fidelity perioperative database, with prediction restricted to information available within the first 60 intraoperative minutes. Among 2,413 waveform-usable cases (180 AKI-positive; 7.46% prevalence), SynerT fell well below strong structured-data baselines, demonstrating that waveform-only temporal modeling is insufficient under strict early constraints. SynerTMM recovered discrimination by incorporating hemodynamic burden summaries and preoperative
covariates, and SynerT-Stack achieved the best overall performance across AUROC, AUPRC, and F1-max. Cross-fitted Platt recalibration substantially corrected calibration defects in both
multimodal variants, and decision-curve analysis confirmed the recalibrated stacked model delivered the strongest net clinical benefit across low-to-intermediate thresholds.
Read the full abstract ▾
All authors Quang Minh Nguyen, Duc Minh Le, Ho Nhat Minh Nguyen, Thuy Quynh Nguyen, Trong Nghia Nguyen
Software Engineering
Who Finishes the Job? A Study of Follow-Up Fixes and Commit Authorship on AI Coding Agent Pull Requests
Wannita Takerngsaksiri, Nhat Duong and Scott Barnett
AI coding
agents now author a large share of pull requests (PRs) merged into popular open-source projects. A merged
agent PR is usually considered finished work; yet, prior studies have reported issues in
agent code after the merge (e.g., code smells and static-analysis issues). However, little is known about how often a merged
agent PR is fixed afterward, and who actually authors the fixing. In this paper, we follow 6,774 merged
agent PRs across five AI coding
agents (OpenAI Codex, GitHub Copilot, Devin, Cursor, and Claude Code) from the AIDev-pop
dataset (open-source repositories with at least 500
stars) into their follow-up fixes, against a baseline of 5,044 contemporaneous human PRs from the same repositories. We link each merge to its candidate fixes, verify every candidate with human annotators and an LLM judge that matches human-level agreement (binary Cohen's Kappa=0.78 against a human-human K=0.77, Direct-fix precision 90%), and attribute the fixing work at the PR and the commit level. Our findings show that (1) merged
agent PRs attract verified fixes at 1.62 times the odds of merged human PRs in the same repositories over the same period of time; (2) 69.6% of verified fixes in
agent merges come from the same
agent; and (3) 76.4% of the verified fix PRs are
agent-authored throughout all commits. These results show that
agents currently largely finish their own job, but their merges still require fixing more often than human merges.
Read the full abstract ▾