Peer Revenue Play credits. Real papers.

Papers

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

All of arXiv

Thursday 10 September

Populations and Evolution

Insights into human evolution from large genetic biobanks

The development of large genetic biobanks with deep phenotyping and whole-exome or whole-genome sequencing is enabling an ever-deeper understanding of human evolution. Here we review recent developments in two main research directions: 1) using the increasing sample sizes to infer extremely strong evolutionary constraint and 2) leveraging biobanks' extensive phenotyping to relate the effects of natural selection on variants to their impacts on traits. In particular, there has been recent progress in gene-specific estimates of the strength of selection acting against loss-of-function mutations, as well as growing evidence supporting the importance of pleiotropic stabilizing selection on traits in shaping patterns of genetic diversity. We also discuss how these findings are, in turn, improving our understanding and interpretation of the genetic associations discovered in biobanks. Throughout, we outline open questions and promising directions for future research.
Read the full abstract

Geophysics

Preemptive Ensemble Forecast Sensitivity to Observations

Estimating the impact of observations newly added to an existing numerical weather prediction (NWP) system requires reintegrating the ensemble forecast from the analysis that assimilates the additional observations, which is laborious and computationally expensive. We therefore propose preemptive EFSO (PEFSO; preemptive ensemble forecast sensitivity to observations), which estimates observation impact without model reintegration by invoking a stronger tangent-linear approximation than that used in EFSO. The estimated impact, however, is merely a scalar quantity at a specified verification time, so it is also valuable to obtain the ensemble forecast in which the remaining additional observations are assimilated after denying the detrimental ones. We refer to this procedure as ADD-SEL, but carrying it out requires reintegration. We therefore propose two methods for updating the ensemble forecast without reintegration as approximations to ADD-SEL, collectively termed ADD-SEL-PRE: one recomputes the ensemble transform matrix (i.e., the Kalman gain), and the other modifies only the innovations at lower cost. Experiments with the Lorenz-96 model examine whether both can be approximated without reintegration. PEFSO estimates observation impact comparable to that obtained by EFSO within the range where the tangent-linear approximation remains valid, even under more practical conditions: an ensemble size of 10, the analysis as the reference state, and only the additional observations available. When additional observations are assimilated after denying detrimental ones, both ADD-SEL-PRE methods update the ensemble forecast, giving forecast errors comparable to those from ADD-SEL, particularly for lead times up to two days, where the tangent-linear approximation is expected to hold.
Read the full abstract

Geometric Topology

Amphichiral Knots: Odd Braid Index and Symmetry Classification in the Three-Braid Case

We study how amphichirality of a knot constrains its braid index and, in the smallest nontrivial case, the combinatorics of its braid words. Using the Dynnikov–Prasolov resolution of the Jones conjecture, we show the braid index of an amphichiral knot is odd. This answers, in the negative, a question of Stoimenow on amphichiral knots of even braid index. We then classify prime amphichiral knots of braid index~$3$. Every amphichiral knot of braid index~$3$ is alternating and admits a minimal $3$-braid representative in a standard form encoded by a word~$c$. Building on the Birman–Menasco classification of closed $3$-braids and the Murasugi normal form, we describe a dihedral action on~$c$ under which the mirror and mirror-reverse operations are realized by a rotation and a reflection. For a standard form whose closure is a prime knot of braid index~$3$, this yields a complete criterion: $\widehat{\beta_c}$ is amphichiral if and only if $c$ is a palindrome or has odd period, together with a determination of the precise symmetry type in terms of these properties and the presence of a non-degenerate flype.
Read the full abstract

Soft Condensed Matter

Self-organisation in hard-soft granular mixtures

Self-organisation of granular systems is a key determinant of their macroscopic behaviour and has been studied extensively in assemblies of hard particles. We use numerical simulations to test this understanding in mixtures of hard and soft particles, focusing on cells, the smallest irreducible loops of the contact network, as structural descriptors. We show that, while the cell statistics display robust qualitative features under isotropic compaction, they depend on inter-particle friction, $\mu$, and soft-particle fraction, $\kappa$. Specifically, (i) the quadron area distributions retain a $\Gamma$ form, albeit with parameters that vary systematically with $\mu$ and $\kappa$. (ii) Predictions of the cell order distribution (COD) by maximising the entropy, without taking mechanical stability into consideration, become increasingly inaccurate at large cell orders. (iii) Irrespective of $\mu$, the normalised cell stress distributions collapse onto one master Weibull form, whose only shape parameter depends weakly on $\kappa$. This suggests a quasi-universal form that may deteriorate slightly at very high fractions of soft particles. (iv) Cells align preferentially along the local major principal stress direction, showing the same coordinated stress–structure self-organisation as in hard particles. The relative robustness of cell statistics to the addition of soft particles suggests that hard and hard–soft granular mixtures can be described by one model.
Read the full abstract

All authors Haoran Jiang, Dominik Krengel, Takashi Matsushima, Raphael Blumenfeld

Combinatorics

Bicoloured-interval and interval-sandwich graphs: two new classes in the tolerance hierarchy

We introduce two new graph classes, the bicoloured-interval graphs and the interval-sandwich graphs, arising from the study of the robust line fitting problem in computer vision. In a bicoloured-interval graph, each vertex is assigned a real interval and one of two colours. Vertices of different colours are adjacent precisely when their intervals intersect, and vertices of the same colour precisely when the centre of one interval lies in the other. Discarding the colouring yields the interval-sandwich graphs, sandwiched between the $50\%$-tolerance graph and the intersection graph of a family of intervals. We prove structural results for both classes, determining which holes, antiholes, trees, and complete bipartite graphs each contains. Consequently, neither class is characterised by finitely many forbidden induced subgraphs. We place both classes strictly between the unit tolerance graphs and the co-comparability graphs in the tolerance hierarchy, separating them from the neighbouring classes. In particular, we construct an infinite family of proper tolerance graphs, each of which is a minimal forbidden induced subgraph for the bicoloured-interval graphs. Finally, we consider the recognition problems for both classes. We show that every graph in either class has a polynomial-size integer representation, and hence that both problems lie in~$\textsf{NP}$.
Read the full abstract

Robotics

Mission Performance: Automatic and Adaptive Race Pace Progression for Autonomous Racing

In this paper, we describe the Mission Performance module implemented for a fully autonomous racing car to automatically manage the longitudinal, lateral, and combined performances, aiming to speedup the laptime progression while assuring safety. Motivated by the difficulty and risks of applying the real-time estimation of the grip to critical modules like the motion planner and controller, the Mission Performance guides these modules adapting their target performance instead of changing the vehicle model parameters. The module is formed by pre-defined progressions to warm up the tires at the beginning of a run. Then, the system continuously monitors safety and vehicle dynamics metrics on a per-sector basis to adaptively reduce, maintain, or increase the performance levels for each sector, progressively converging toward the maximum allowed value. The solution's effectiveness is demonstrated on the EAV-25, a fully autonomous Dallara Superformula, at the Yas Marina Circuit during the Abu Dhabi Autonomous Racing League (A2RL) Season 2.
Read the full abstract

All authors Giovanni Lambertini, Matteo Pini, Nicola Musiu, Ayoub Raji, Francesco Iacovacci, Marko Bertogna

Theoretical Economics

Measuring Economic Preferences in the Presence of Noise: The Connections Between Choices and Valuations

Past research highlights failures of "procedural invariance" when measuring economic preferences using choices versus valuations. We reassess these failures by examining theoretical connections between choices and valuations when preferences are stable but measurements are noisy and individuals are heterogeneous. Even under strong assumptions governing noise and heterogeneity, stability does not generally imply identical measurements. We develop new tests of stable preferences in conjunction with various ancillary assumptions about heterogeneity and noise. We implement these tests using existing data to understand if, in the domain of risk preferences, choices and valuations truly differ and to provide quantitative assessments of any deviations. Limiting to the types of data used in the prior literature, we rarely reject the null of stable preferences. With richer data linking individual choices and valuations and structural assumptions, we find evidence of instability which differs qualitatively from the received wisdom that choices implicate greater risk aversion than valuations.
Read the full abstract

All authors Ted O'Donoghue, Charles Sprenger, Po Hyun Sung, Ben Wincelberg

Combinatorics

Cayley-graph density of Thompson's group F: local deletions and finite-window bounds

Let $\Gamma$ be the Cayley graph of Thompson's group $F$ with its standard generators. Belk and Brown constructed finite marked-forest subgraphs of limiting density $7/2$, and Guba showed that deleting certain low-degree vertices gives density greater than $3.5004$. We refine this approach in two directions. First, a deterministic interval-deletion rule followed by root-sensitive triple and pair deletions yields finite induced subgraphs with \[ \operatorname{dens}(\Gamma)>3.50074529. \] The interval rule is evaluated by a nine-state recurrence, while the simultaneous deletion conditions created by split and merge operations are computed exactly by a common-suffix first-passage argument. Second, we determine the joint distribution of the categories of a tree and its two children and reduce optimization over every fixed category-window retention rule to a weighted densest-subgraph problem. An explicit edge-allocation certificate shows that, for every fixed $a\ge0$, the optimal limiting density of a rule depending on the window $[-a,1]$ is exactly $7/2$. Thus arbitrary finite left context together with one right-hand category does not improve the Belk–Brown limit, whereas the root-sensitive whole-segment construction does. All constants are explicit elements of $\mathbb Q(\sqrt3,\sqrt{2\sqrt3-1})$.
Read the full abstract

Quantum Physics

Dynamical qubit cooling via quasi-particle transport

Cooling a quantum state presents a challenge as the external control in active protocols can induce detrimental heating in the system. Passive cooling processes offer a valuable alternative: Here, we present a simple quench protocol to cool a set of qubits with access to an ancillary spin chain. We show that excitations can be transported out of the qubits as quasi-particles, taking the form of magnons in a ferromagnetic ancilla and triplons in a dimerized antiferromagnetic ancilla. This quasi-particle transport mechanism enables cooling towards both individual and highly entangled two-qubit target states. We demonstrate that this cooling mechanism remains robust starting from thermal states and with finite quench ramp times.
Read the full abstract

Human-Computer Interaction

"People can change, and patterns can be broken": Contextualizing Tradeoffs in Automated Decision-Making Systems

Automated decision-making (ADM) systems are increasingly deployed in domains such as mortgage lending, prison sentencing, health insurance coverage, and hiring. Designing a responsible ADM system in such high-stakes domains requires ensuring privacy protection, fairness across demographic groups, and robustness against adversarial manipulation. However, prioritizing one of these objectives comes at the cost of another, forcing a choice as to which tradeoff to accept in a deployment. These tradeoffs explicitly or implicitly impact the life, safety, and fundamental rights of the people in a society, and thus, the perceptions and priorities of this population are needed before we can produce appropriate solutions. To this end, we conducted a quasi-experimental study (N = 777) in which participants evaluated four decision-making scenarios with controlled tradeoffs. Participants significantly preferred human decision-making (HDM) over ADM in three of four scenarios, emphasizing the value of human judgment, contextual understanding, and the ability to incorporate non-quantifiable factors. Furthermore, in terms of tradeoffs, our findings not only show that participants' preferences are highly context-dependent, but also that their perception of a specific objective, fairness, extends beyond formal definitions. Participants interpret fairness through multiple lenses, including privacy risks and susceptibility to manipulation, and view unfair or manipulated outcomes as failures of accuracy. Overall, our findings highlight the importance of context-aware and human-centered approaches when designing and governing ADM systems in high-stakes situations. Rather than purely technical objectives, it is essential to evaluate ADM systems based on how their tradeoffs align with specific expectations within a given domain, as well as with societal values and perceptions of harm and fairness.
Read the full abstract

All authors Rabeya Bosri, Anna Harbluk Lorimer, Afrida Hossain, Vasisht Duddu, Bailey Kacsmar

Artificial Intelligence

Robust Prototypical Networks for Few-Shot Sensor Fault Diagnosis

Industrial fault diagnosis often operates with only a handful of labeled fault examples, making few-shot learning attractive for sensor monitoring. Standard prototypical networks are simple and effective; however, their class prototypes may become unstable in the very-low-shot regime because each decision relies on a small support set. We propose Multi-Episode Prototypical Networks (MEPN), which aggregate prototypes from multiple disjoint support episodes and use their mean as the final class representative, reducing prototype variance without changing the encoder architecture. We evaluate MEPN on the DeFACTO sensor dataset using five-way fault classification with synthetic bias, drift, spike, and noise faults injected into real industrial measurements. Over 100 independent runs, MEPN reaches \SensorOneShotGcpn% in the per-episode one-shot setting ($K\!=\!1$ shot, aggregated over $N_{\text{agg}}\!=\!10$ support episodes), substantially above single-episode baselines. Under an equal 10-sample support budget, MEPN and ProtoNet at $K\!=\!10$ are statistically indistinguishable, confirming prototype accumulation as the mechanism rather than superior fixed-budget learning.
Read the full abstract

Artificial Intelligence

T-GADE: Thermodynamical Generative-AI-Driven Evolution of LLM Artifacts

Introducing large language models (LLMs) into evolutionary computation enables semantic generation of candidates, but the classical problem of maintaining population diversity remains. Inspired by the thermodynamical genetic algorithm (TDGA), we propose T-GADE (Thermodynamical Generative-AI-Driven Evolution of LLM Artifacts), an evolutionary computation framework extended through LLMs. T-GADE keeps the notions of a population and genetic operators, takes structured artifacts generated by an LLM as individuals, and implements crossover and mutation as LLM operators. It further introduces the HoxLM, an LLM that attempts self-repair of an individual after evaluation and before survival selection; two update schemes, generational and steady-state; and Fermi-type and Bose-type occupancy rules that define solution identity and whether duplicates are permitted. Survival selection is unified as minimization of the free energy $F_T=\langle E\rangle-TH$, and the selection temperature $T$ explicitly controls the balance between quality and diversity. We tested an implementation example aligned with Evolution of Heuristics (EoH) on the online bin-packing task, where the excess is the bin count in excess of the volume lower bound, relative to that bound. The HoxLM-equipped configuration reached lower excess with fewer total LLM calls than EoH. In the preregistered comparison at 400 total calls it won 9 of 10 seed-paired runs, with a median sealed excess of 0.836% against 3.912%, and it was also better in the mean and in the worst case. Temperature controls the spread of the retained population under both update schemes, and a small positive temperature improves training quality over the collapsed $T=0$ point in the generational Bose-type scheme. The evolved heuristics are shown as readable scoring rules.
Read the full abstract

2609.12286 · 10 Sep · v2: 12 pages, 4 figures, 6 tables. Presents T-GADE as a general diversity-aware framework for LLM-based evolutionary computation with an EoH-aligned implementation example, reports a preregistered seed-paired comparison in which the host-equipped configuration reaches low excess with fewer total LLM calls than EoH, and treats the selection temperature as an explicit diversity control

Robotics

AnchorVLN: Geometry-Anchored Vision-Language Grounding Reasoning for Open-Vocabulary Navigation

Vision-Language Navigation (VLN) in unseen indoor environments is useful in real-world robotics, where an agent must follow natural-language instructions, locate objects, and answer spatial questions without a pre-built map or fixed object vocabulary. Multimodal vision-language models (VLMs) provide strong open-vocabulary grounding and zero-shot reasoning, but struggle to emit reliable metric quantities such as range, bearing, and comparative spatial relations directly from images. Existing approaches address this by folding geometry into hand-engineered pipelines or asking models to output waypoints, requiring changes to the control stack for different robots, tasks, or vocabularies. We introduce AnchorVLN, an open-vocabulary VLN system built on a simple rule: the VLM proposes semantics; geometry decides metrics. It is realised as EMBODIED-NAV-MCP, a Model Context Protocol (MCP) server driven by a VLM agent through a compact set of callable tools. Since no tool accepts distance in metres or bearing in radians, the schema enforces the semantic-geometry boundary without modifying the downstream autonomy stack. We benchmark both tasks of the CMU Vision-Language Navigation Challenge 2026: 30 instruction-following questions over 15 scenes and a frozen 45-question object-reference set. The full system achieves 64.4 percent on instruction following, dropping by 13.3 percentage points without controller modeling (t = 2.77). On object reference, geometric anchoring clears the challenge overlap threshold on 10 of 45 questions, versus 0 of 45 for direct coordinate estimation, reducing median center error from 3.37 m to 2.48 m.
Read the full abstract

All authors Long Giang Vu, Chengkai Yao, Yuxin Liu, FNU Aryan, Rajath Chandrashekar Aralikatti

Quantum Physics

A Simple Quantum Linear-System Solver via Dissipation

Dissipation has been recently demonstrated as a powerful primitive for designing quantum algorithms. We apply this viewpoint directly to linear-system solving, $Ax=b$. We construct a simple purely dissipative Lindbladian whose unique fixed point encodes the linear-system solution. We prove dimension-independent trace-distance mixing in $\Theta(\kappa^2\log(1/\eps))$ time. We then show how to run this Lindbladian on digital quantum computers via collective block encoding and Lindbladian simulation, resulting in an $O\!\left(\kappa^2\log(1/\eps) \frac{\log(\kappa/\eps)}{\log\log(\kappa/\eps)}\right)$ query complexity for both $U_A$, the block encoding of $A$, and $U_b$, the $|b\rangle$ state preparation unitary.
Read the full abstract

Classical Analysis and ODEs

An Elementary Proof of the Hambly-Lyons Uniqueness Theorem

We give a self-contained proof, in the bounded variation setting, of the Hambly–Lyons uniqueness theorem, which states that (total) signature identifies the path up to tree-like equivalences. The argument is organized around two key geometric observations. First, tree-like paths have trivial signature because factorization over a loop in a tree $\tau:[0,1]\to T$ is preserved under signature lifts, which follows from an elementary property of planar curves. Second, a path with trivial total signature contains a nontrivial subpath with trivial total signature (the sub-interval lemma). This is proven by applying a winding-number argument to a two-dimensional projection of the signature lift. Collapsing all trivial-signature sub-intervals then defines a compact metric tree $T$ through which the original path factors by virtue of the sub-interval lemma.
Read the full abstract

Numerical Analysis

Adaptive local representations for Helmholtz Trefftz discontinuous Galerkin methods

We study the selection and stable realization of local approximation spaces in Trefftz discontinuous Galerkin discretizations of the Helmholtz equation. A scaled Cauchy-trace inner product places plane waves and Fourier-Bessel functions in a common geometry: Fourier-Bessel modes are orthogonal with explicit weights, while the same weights determine the circulant spectrum of an equispaced plane-wave trace Gram matrix. This separates amplitude scaling from genuine trace-rank loss and yields an exact best-approximation identity for mixed plane-wave–Fourier–Bessel spaces. With complex plane-wave angles, the unresolved modal tail is an exponential sequence, so propagating and evanescent components can be identified by the same ESPRIT/variable-projection procedure. We prove exact recovery and a perturbation estimate for the recovered angles and, under the standard PWDG quasi-optimality bound, transfer these perturbations to the DG error. Trace-Riesz orthonormalization is then separated from a graph–Riesz normalization of the assembled operator. Numerical experiments verify the identities, recover sparse ray fields to roundoff, and resolve a propagating-to-evanescent transition without a prescribed critical angle.
Read the full abstract

Number Theory

On the variation of slopes over moduli stacks of Galois representations

We determine how p-adic slopes of modular forms vary as their associated Galois representations modulo p vary over moduli of Galois representations. As a consequence, we show that the authors' ghost conjecture recently proven by Liu, Truong, Xiao, and Zhao is "stable under direct sums".
Read the full abstract

Econometrics

Simultaneous Clustered Orthogonalization

Identification in vector autoregressions involves two distinct choices: how extensively to orthogonalize shocks and whether orthogonality is imposed sequentially or simultaneously. Generalized identification imposes no orthogonality; Sims (1980} imposes full orthogonality sequentially; Francis et al. (2026) impose full orthogonality simultaneously; and Buchwalter et al. (2026a) provide clustered partial orthogonalization sequentially. We fill the remaining case by developing simultaneous clustered orthogonalization (SCO). SCO preserves contemporaneous dependence within economically meaningful clusters while imposing orthogonality across clusters jointly, thereby eliminating dependence on cluster ordering. We formulate the associated correlation-maximizing identification problem and show that, in correlation space, it reduces to a quadratic matrix equation. This yields a closed-form solution, and we prove that the associated identification matrix is the unique global maximizer. SCO is order- and scale-invariant and nests generalized identification and full simultaneous orthogonalization as special cases, yielding a flexible family of structural decompositions indexed by the number and composition of clusters.
Read the full abstract

General Relativity and Quantum Cosmology

Dynamics for Spin-$1/2$ Particles in Einstein-Gauss-Bonnet Gravity II: Non-Relativistic Case

In this work, I investigate the non-relativistic quantum dynamics of spin-1/2 particles in Einstein-Gauss-Bonnet (EGB) gravity and establish a direct connection between higher-curvature corrections, fermionic dynamics, and the phenomenology of compact objects. Starting from the Dirac Hamiltonian in a static, spherically symmetric EGB spacetime, we perform a Fold-Wouthuysen transformation and derive the effective Hamiltonian, including relativistic kinetic, gravitational, spin-orbit, and higher-curvature contributions. Heisenberg equations are then used to obtain the dynamics of velocity, force, and spin, revealing explicit EGB corrections for both translational motion and spin transport. In particular, the spin-orbit sector induces a modified precession frequency whose fractional deviation from general relativity scales as $\delta_{\Omega}=-4(\xi/M^{2})(M/\rho)^{3}$, providing a clear dimensionless signature of the Gauss-Bonnet coupling. Through Ehrenfest's theorem, we also establish the correspondence between the dynamics of quantum operators and their semiclassical gravitational limit. As an astrophysical application, we consider the stellar-mass black hole A0620-00 and show that prospective relative sensitivities in spin precession on the order of $10^{-3}$ to $10^{-4}$ can probe Gauss-Bonnet couplings in the range of approximately $10^{6}$ to $10^{8}\,{\rm m}^{2}$, depending on the orbital radius. This result identifies fermionic spin precession as a complementary channel for testing gravity with higher-curvature corrections and provides a quantum-mechanical framework connecting modified gravitational dynamics to precision phenomenology in strong-gravity regimes.
Read the full abstract

Machine Learning

Amortized Low-Rank Adaptation for Model-Based Reinforcement Learning

World models let agents plan by predicting the consequences of their actions, but changes in the environment can make them inaccurate. We study the problem of adapting a world model to an unknown test-time environment, drawn from a known environment family, using only a few episodes of interaction. Existing approaches trade off computational cost against expressivity, i.e., the range of models a method can produce. For example, in-context learning is computationally cheap but limited in expressivity, and gradient-based adaptation is expressive but computationally expensive. We present CLAW (Context-conditioned Low-rank Adaptation of World models), which addresses this tradeoff by using a hypernetwork to generate low-rank (LoRA) adapters at test time. During pretraining, we simulate adaptation to a variety of environments and jointly train the hypernetwork and base world model. At test time, we freeze the base model and use a forward pass of the hypernetwork to generate adapters from a small batch of test-time transitions. We evaluate CLAW in locomotion and manipulation environment families that vary in dynamics, embodiment, and reward. We show that, using only seconds of test-time data, CLAW outperforms gradient-based adaptation and in-context learning during online adaptation. We also show that CLAW avoids overfitting in data-scarce regimes, that its advantage comes from the expressive adapters rather than context conditioning, and that pretraining the hypernetwork jointly with the base model outperforms training it post hoc.
Read the full abstract

Machine Learning

Reinforcement Learning over Patient Trajectories for Clinical Reasoning in EHR Foundation Models

Electronic health record (EHR) foundation models trained on longitudinal patient trajectories have demonstrated strong performance across diverse clinical prediction tasks. However, their clinical reasoning capabilities remain constrained by next-token prediction on limited and incomplete EHR data. To address this, we propose a reinforcement learning (RL) fine-tuning framework that treats EHR foundation models as generative policies over patient trajectories. We formulate common clinical prediction problems (e.g., hospital readmission) as event-conditioned, time-windowed reasoning tasks. We then design time-aware, rollout-sensitive rewards to account for finite rollout lengths and temporally inconclusive outcomes. We find that RL fine-tuning consistently improves over pre-trained backbones and strong baselines. Notably, it enables smaller models to match larger pre-trained models in data-limited regimes and induces positive transfer across tasks. Further analysis shows that RL fine-tuned models generate trajectories with stronger structural and semantic alignment to ground truth and greater downstream utility.
Read the full abstract

All authors Yuxin Xiao, Sheng Zhang, Chandan Singh, Tristan Naumann, Hoifung Poon, Jianfeng Gao, Xiaodong Liu

Operating Systems

AKTS: Sub-Microsecond Kernel Policy Switching for Language-Model Agents

GPU-backed LLM servers multiplex interactive requests with background batch work on the same CPUs; a fixed kernel policy serves one objective and loses the other, so agentic OS control needs to switch scheduler behavior as the workload changes. The hard part is applying the switch safely and fast enough for the kernel: scheduler events occur every 1-10 $\mu$s, and any code running there must satisfy the eBPF verifier. Scalar knobs are fast but limited, while generating eBPF policy code puts compilation, verification and possible rejection on the runtime path. We present AKTS, which verifies a policy library once, at load time, and reduces the agent's runtime action to writing an integer index into an in-kernel array of preverified policies, resolved by a tail call. Because the agent emits an index rather than code, verifier failure is not a runtime outcome. On Linux 6.14, AKTS applies a policy switch in 920 ns (p50); makes an invalid index inert across 60,217 invocations on an attached scheduler; and switches policies in a vLLM workload to capture 97% of a throughput policy's batch work while matching a latency policy's burst response.
Read the full abstract

Space Physics

Low-frequency Intermittency and Structures in the Solar Wind at 1 au

Intermittency in the solar wind is commonly studied within the inertial and dissipative ranges, where scale-dependent magnetic field distributions become increasingly non-Gaussian toward smaller scales until dissipation becomes important. Conversely, whether Gaussianity is recovered at large scales remains unclear. To address this, we systematically examine magnetic field increment kurtosis over scales from 1 minute to 1 year using more than two decades of in situ observations from NASA's Wind and ACE spacecraft. We find that although kurtosis tends toward Gaussian value of 3 near the correlation scales, it generally remains elevated (super-Gaussian) at larger scales. Kurtosis also varies substantially over time, with pronounced super-Gaussian intervals during high solar activity phases and sporadic sub-Gaussian intervals primarily in the radial component. These results provide evidence for large-scale intermittency in the solar wind, potentially arising from mixing of different solar wind streams and nonstationary driving of solar sources.
Read the full abstract

All authors Jiaming Wang, Francesco Pecora, Yan Yang, William H. Matthaeus

Geophysics

Ensemble Forecast Updates without Model Re-integration Based on Ultra-rapid Data Assimilation: Idealized Experiments with a Heavy Rainfall Case

Observations at high frequency have become dramatically more abundant with recent technological advances. Ultra-rapid data assimilation (URDA) has been proposed to exploit them, frequently updating ensemble forecasts at lower cost without re-integrating the forecast model. In this study, we examine the applicability of URDA to a realistic numerical weather prediction (NWP). Specifically, we conducted idealized experiments for the heavy rainfall event of August 2021, using the regional atmospheric model Scalable Computing for Advanced Library and Environment (SCALE-RM). In the experiments, pseudo-observations emulating the Automated Meteorological Data Acquisition System (AMeDAS) were assumed to become available every 10 min. The results show that the threat score for hourly precipitation is generally improved in the forecasts updated by URDA relative to the pre-existing baseline forecasts, through the ensemble-based error covariance. Furthermore, for sea level pressure, temperature, relative humidity, and winds, the root mean square error of the URDA-updated forecasts is reduced relative to that of the baseline forecasts as the forecasts are successively updated by assimilating additional observations. These results indicate the potential of URDA to operate effectively with a realistic NWP model.
Read the full abstract

All authors Fumitoshi Kawasaki, Kenta Kurosawa, Atsushi Okazaki, Shunji Kotsuki

Human-Computer Interaction

Synthetic TLX: Forecasting Human Workload Using Agent Simulation

Assessing human workload for technology-mediated tasks helps prevent task failure caused by poor technology design. Traditionally, workload is assessed retrospectively using the NASA Task Load Index (TLX) after humans complete a task. What if we could forecast workload before a human attempts a task using agent simulation? We introduce Synthetic TLX, a new paradigm for proactive workload estimation that predicts NASA TLX scores for a given task, unlocking novel interaction opportunities and evaluation methods. To understand its viability, we conducted three experiments comparing human and agent-generated scores to evaluate where they align and diverge. We found agent estimates align with human scores particularly when prompted with a human persona and active task simulation. However, agents and humans diverge in the sources of workload they are sensitive to. Based on our findings, we present three applications to showcase Synthetic TLX's potential and discuss the future of workload-aware human-AI interaction.
Read the full abstract

All authors Tzu-Sheng Kuo, Carrie J. Cai, Meredith Ringel Morris, Michael Terry

Human-Computer Interaction

Tact: A Zero-Cost, Browser-Based Pipeline for On-Demand Tactile Braille Storybooks

Children who learn braille have few illustrated books to read with their fingers, and almost none about the things they ask for. Existing approaches either adapt a published picture book by hand, with a specialist in the loop, or generate a tactile image without the braille page around it. We present Tact, an open system that turns a one-sentence idea into a printable page of braille with a raised picture, in Italian or English, on a consumer 3D printer, with no account and no cost beyond the plastic. Its central idea is to treat the physical page as a constraint on generation: the cell budget of a fixed tactile layout decides how long a story the language model is asked to write, and the story is measured and requested again until it fills two cards. The page itself follows requirements traced to braille standards, the tactile-graphics literature, and formative feedback from a teacher of students with visual impairments and a blind reader who printed and read a page. In an evaluation of 60 stories generated live, every story fit exactly two cards and 57 filled the second card to at least 11 of its 13 rows; a final targeted round recovered 5 of 6 retries of the remaining cases. The picture chosen matched the story's subject on 105 of 105 illustrated pages, against 63 for first-match keyword selection. Uncontracted Italian and English braille, and contracted English produced by liblouis compiled to WebAssembly, matched native liblouis cell for cell on 106 texts, and the interface showed no WCAG 2.2 AA violations. Cards are 150 mm square on a 2 mm plate, printed standing up, and take about 57 g of PETG; test prints are under evaluation.
Read the full abstract

Statistics Theory

A Sharp Signal-to-Noise Threshold for Quasi-Maximum Likelihood Breakpoint Estimation

We establish a sharp pathwise signal-to-noise criterion for quasi-maximum-likelihood (QML) estimation of a dominant breakpoint in the second-moment structure of a multivariate time series: the QML estimator is consistent whenever the between-regime contrast exceeds the within-regime fluctuation by an explicit factor, and below this threshold global recovery can fail. Two innovations drive the result. First, the framework is pathwise: no stochastic model is imposed on the data. Second, the log-determinant objective carries an additive ridge regularization: it removes the endpoint boundary layers, so no trimming of the candidate set is required, and tuning the ridge weakens the consistency condition. As an application of the main theorem, we establish consistency of QML breakpoint estimation in pervasive factor models whose dimension diverges with the sample size, for completely general error terms – not necessarily independent, idiosyncratic, or even random.
Read the full abstract

Information Retrieval

Recommendation Retrievers Need Verifiers: Universal Generative Reranking for Sequential Recommendations

First-stage recommenders in multi-stage systems produce a ranked candidate list from which a limited prefix is forwarded to downstream rankers. Because each forwarded item must be processed by more expensive ranking stages, this shortlist cannot be arbitrarily large. The first-stage objective is therefore high coverage of relevant items within the forwarded prefix, commonly measured by Recall@$k$. A relevant item may be available deeper in the retrieved list but absent from the shorter prefix that is actually consumed. This paper studies post-hoc verification for promoting such candidates into the consumed shortlist without retraining or replacing the retriever. We introduce a lightweight generative verifier for retrieval models. Given a retriever state and a candidate item, the verifier scores the item through the likelihood of its identifier tokens. It is trained post hoc with next-token cross entropy, requires no sampled negatives or candidate pool during training, and scores only the retriever's top-$K$ candidates at inference. The interface is minimal: the retriever supplies a query state and candidate items, and the item representation can use any fixed tokenization. Across Amazon product recommendation and YaMBDa music recommendation, the same verifier training recipe improves Recall@10 for SASRec, GRU4Rec, NextItNet, and MiniOneRec. Ablations show that the improvements are not explained solely by injecting item-content features into the retriever, supporting verification as a post-hoc output-side adaptation mechanism.
Read the full abstract

All authors Benyu Zhang, Qiang Zhang, Rui Li, Qunshu Zhang, Devansh Tandon, Neeraj Bhatia

Optics

Inverse reconstruction of dissipative Kerr soliton interactions

We reconstruct the pairwise interaction potential between dissipative Kerr solitons from the linear stability spectrum of a perfect soliton crystal solution of the Lugiato–Lefever equation. The solitons form an overdamped lattice whose positional eigenvalues encode the pairwise-force derivative, mirroring the extraction of microscopic interactions from phonon or relaxation spectra in condensed-matter systems. The method extends beyond quadratic dispersion and assumes no functional form for the interaction. Our theory predicts a novel state -the soft soliton crystal- which has vanishing stiffness and a diverging positional relaxation time, and also explains experimentally observed soliton steps as arising from a sign reversal of the pairwise-force derivative.
Read the full abstract

All authors Carlo Silvestri, Panayotis G. Kevrekidis, Stephane Coen, C. Martijn de Sterke, Antoine F. J. Runge

Computation and Language

EAR: Entity-Aware Partitioning Approach for Retrieval-Augmented Generation Development

Retrieval-augmented generation (RAG) can improve knowledge-intensive question answering, but the first design choice is easy to overlook: how should the source corpus be partitioned into retrievable units? Fixed-size chunks often return long passages whose relation to the question is only implicit. We introduce EAR, an Entity-Aware Partitioning approach for multiple-choice question answering (MCQA). EAR extracts normalized surface anchors from the question, answer options, and corpus; retrieves local windows around matching corpus anchors; and can attach a larger parent passage through an extractive summary. We evaluate EAR on a cleaned Massive Multitask Language Understanding (MMLU)-style subset of 153 questions selected by an automatic corpus-support heuristic and using decontaminated public textbook text. Across same-protocol top-k = 3 and top-k = 8 sweeps with Mistral, Gemma, and DeepSeek, EAR entity-window reduces retrieved words by 37.5-40.2% relative to chunks. Observed accuracy changes are +5.2, +1.3, and -3.9 points at top-k = 3, and +5.9, -3.3, and -4.6 points at top-k = 8; none of the entity-window differences is statistically significant. The scoped contribution is methodological: EAR provides a compact and inspectable retrieval unit, while its rule-based anchor extractor remains domain-specific and requires separate validation before transfer.
Read the full abstract

All authors Cenab Batu Bora, Oylum Alatl{\i}, Sebnem Bora, Oguz Dikenelli