Peer Revenue Returns subject to peer review

Papers

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

All of arXiv

Tuesday 1 September

Dynamical Systems

Countable IET Models and Defect Sets for Interval Translation Maps

Interval translation maps are piecewise translations for which the images of distinct continuity intervals may overlap. We study when their measured dynamics can be represented by finite or countable interval exchange transformations. First, we give a direct entropy-based proof of the known fact that an interval translation map is invertible almost everywhere with respect to every nonatomic invariant probability measure. The proof uses a polynomial upper bound for the complexity of the natural branch coding. We then use a consequence of a theorem of Arnoux, Ornstein, and Weiss: every nonatomic measure-preserving automorphism of a standard probability space admits a countable interval exchange transformation model. Applied to the almost-everywhere invertible core, this gives an abstract countable interval exchange transformation model for every measured interval translation map. We next study the canonical order-preserving coordinate defined by the distribution function of the invariant measure. For each branch, we introduce a positive defect measure recording the excess measure carried by its direct image. Its push-forward to the distribution coordinate gives a canonical cut set such that, away from this set, the induced map is locally a translation. Finite defect support yields a finite interval exchange transformation model, while a Lebesgue-null defect cut set yields a countable model. Under branchwise nonsingularity, the defect support is the closure of the cuts arising from active gaps in the support, giving an equivalent geometric characterization of the finite case. Finally, for a self-similar infinite-type Bruin-Troubetzkoy map, we compute the canonical defect cut set explicitly. It is countably infinite and Lebesgue-null, yielding a genuinely countable, non-finite interval exchange transformation model.
Read the full abstract

Software Engineering

Barriers to Using Static Application Security Testing (SAST) Tools: A Literature Review

Developers face a challenging problem with no clear solution. Modern software breaches can wreak havoc on businesses and individuals alike. With code vulnerabilities being a leading cause, securing applications must be a priority for developers. Static Application Security Testing (SAST) has the potential to harden applications by assisting in the identification and resolution of security vulnerabilities. Despite this, many development teams have not adopted SAST tools into their environment. In this paper, we survey the recent literature to uncover why some developers are apprehensive towards SAST and identify what specific problems they encounter when using it. We found a variety of usability problems developers face when using SAST. Some are inherent of the tool and ultimately require some level of developer investment while others are tool shortcomings that SAST tool creators must address. Ultimately, we argue that in order to drive widespread adoption and consistent SAST usage, developers will need to embrace that some investment is required. Simultaneously, developers will be more likely to integrate SAST tools into their workflows if the creators of SAST tools simplify many aspects related to tool usage. Surmounting the primary obstacles preventing the adoption of SAST requires full consideration of both the technical and human factors.
Read the full abstract

Adaptation and Self-Organizing Systems

Arithmetic of the sync basin for pulse-coupled oscillators

A population of $N$ identical pulse-coupled oscillators ultimately settles into one of two outcomes: full synchrony or a state of co-existing synchronized clusters. We show that which outcome occurs is controlled by the prime factorization of $N$. At the critical charging curve — linear, the boundary between the synchronizing and clustering regimes — the synchronization basin acquires exact arithmetic structure. The synchronization probability is $\Psync=A_{N,1}/N^N$ for all $N$, where $A_{N,1}$ satisfies an exact recurrence relation. For prime $N$, $A_{N,1}=N^N-1$ giving the closed form $\Psync=1-1/N^N$; for composite $N$, the observed asymptotic scaling is $1-\Psync\sim C_m N^{-(m-1)}$, where $m$ is the smallest prime divisor. The result adds a new member to the atlas of exotic basin geometries: alongside fractal, riddled, and tentacled basins, we now have a basin that is arithmetic.
Read the full abstract

Cryptography and Security

Private Computation Space: Experience with Trusted Multi-Cluster Federated Learning for Agriculture

Artificial Intelligence has shown to help improve agricultural practices, yet adoption remains limited: 69% of U.S. farmers have privacy concerns with sharing their data, and these concerns must be addressed before adoption is widespread. While Federated Learning has been demonstrated to protect privacy at scale for other sectors, deploying a system for agriculture comes with its own set of challenges; the problem necessitates a system that can protect farmer data and identities while preserving model utility, runs on commodity hardware, and is resilient to fragile rural infrastructure. To address these concerns, we introduce the Private Computation Space (PCS), a deployed, open-source Machine Learning system to provision and process farmer data securely. We design a system tailored to an agricultural setting, with multi-cluster orchestration for reliability in rural areas with asynchronous Federated Learning (FL), Differential Privacy (DP), and Trusted Execution Environments (TEEs), to allow farms to participate in the framework while keeping their data private. We evaluate the system on two deployed workloads: monitoring nitrogen with living plant sensors in NY for six months and predicting evapotranspiration from weather stations in CA for ten months. Our evaluation finds a Dice Similarity Coefficient (DSC) of 0.71 and $R^2$ accuracy of 0.84 for the respective workloads, improving the worst single-site model accuracy by 22.4% and 9.1%, respectively, while preserving privacy.
Read the full abstract

All authors Shuangyu Lei, Muhammad Salman Abid, Jacob Belding, Sam Mosher, Manushi B. Trivedi, Shivranjani Baruah, Liam Wickes-Do, Andrew Anderson, Braulio Dumba, Alyssa Whitcraft, Ritvik Sahajpal, Sijin Li, Kelly Robbins, Michael Gore, Margaret Frank, Steven Wolf, Liz Jones, Abraham Stroock, Kaitlin Gold, Hakim Weatherspoon

Information Theory

Solomonoff Induction and Singular Integrals

The Solomonoff distribution $M$ assigns an a priori probability to a finite binary string $z$ by summing over all programs whose output begins with $z$, weighting a program of length $\ell$ by $2^{-\ell}$. Thus, likely strings are those with many short explanations. This appears to be a purely discrete notion of complexity. However, Riemann sums are also countable. We show that the sum defining $M$, suitably reorganised, contains Riemann sums approximating the Bayesian evidence of any computable statistical model. These evidence integrals are singular integrals whose asymptotics are governed by Singular Learning Theory and, through it, by invariants of algebraic geometry. Concretely, for every computable Bayesian model, we construct a single monotone Turing machine whose induced semimeasure agrees with the evidence $Z_n$ up to a uniform multiplicative constant. If the model also satisfies the hypotheses of Watanabe's free-energy asymptotics, then a sample $X^n=X_1\cdots X_n$ drawn i.i.d. from the true distribution satisfies $-\log M(X^n) \le nL_n(w_0)+\lambda\log n-(m-1)\log\log n+O_{\mathbb{P}}(1)$, where $L_n$ is the empirical loss, $w_0$ is an optimal parameter, $\lambda$ is the learning coefficient, and $m$ is its multiplicity. Thus the learning coefficient, a geometric measure of model simplicity, appears within the Solomonoff distribution as the coefficient of $\log n$ in an upper bound on code length.
Read the full abstract

Instrumentation and Methods for Astrophysics

Amortized Opacity Marginalization Improves C/O Interval Calibration for Brown-Dwarf Retrievals

Molecular line lists disagree at a level that measurably shifts retrieved C/O, so retrievals conditioned on a single list omit a dominant systematic from their error bars. Classical marginalization re-runs a full retrieval under K opacity realizations per object. We move that cost into simulation: while generating the training set we randomize opacities (smooth multiplicative Fourier fields in log(nu) plus a discrete line-list mixture, under a deliberately broad prior informed by measured inter-list differences), so a neural posterior estimator learns an already-marginalized posterior. On perturbed held-out spectra (n=2048) a standard NPE covers nominal 90% C/O intervals 43.3% of the time against 82.4% for a compute-matched marginalized network; three-seed control ensembles reach 93.2% on clean data but 61.4% under perturbation, against 89.4% marginalized, isolating the deficit as opacity-induced (TARP joint expected-coverage deviation 0.005 against 0.035 for a single control network). An axis ablation traces the gain to the continuous field (45.4% for the discrete swaps alone, 79.3% for the field alone). In simulation, real JWST bad-pixel masks with median filling cut the marginalized ensemble's C/O coverage from 0.894 to 0.683; an ensemble retrained with mask augmentation holds 0.918 under the full deployment preprocessing (n=1024). We apply the networks to one pipeline-combined JWST/NIRSpec G395H spectrum each of 14 late-T and Y dwarfs. The recommended mask-augmented ensemble returns C/O medians between 0.24 and 0.62 and contains the Hood et al. (2024) value of 0.53 for the T8 benchmark 2MASS J0415-0935, which every fixed-opacity control network excludes. With 90% intervals spanning about 0.43 of the prior width, this is a consistency check, not a validation.
Read the full abstract

Computer Vision

Uncovering Understanding-Generation Synergy in Native Unified Multimodal Models: From Representation, Task to System

While unified multimodal models (UMMs) jointly perform visual understanding and generation within a single model, functional unification does not guarantee learning synergy: the two objectives may reinforce each other, compete for capacity, or merely coexist. We investigate their relationship at the representation, task, and system levels in a controlled, structurally native setting without pretrained vision priors. At the representation level, we find that each objective provides useful signal to the other: generation enriches the visual features learned for understanding, while understanding strengthens vision–language alignment for generation. However, when both objectives are forced through the same computation path, one tends to dominate. A task-decoupled architecture that specializes conflicting visual computation while preserving semantic interaction avoids this asymmetric degradation. At the task level, through three case studies, we find positive bidirectional transfer when understanding and generation tasks rely on shared knowledge. At the system level, we show that an end-to-end UMM outperforms a matched planner–executor pipeline on complex tasks that explicitly require both image understanding and generation. Together, these results show that the value of UMMs extends beyond a unified interface: appropriate specialization, shared task knowledge, and end-to-end optimization can turn coexistence into synergy.
Read the full abstract

All authors Penghao Wu, Haiwen Diao, Weichen Fan, Lewei Lu, Dahua Lin, Ziwei Liu

Quantum Physics

Observable- and state-selective prethermalization and bounds on prethermal lifetimes

Prethermalization describes long-lived intermediate regimes that precede equilibrium and can dominate experimentally accessible dynamics. Here, we show that a separation of spectral energy scales, despite giving rise to a hierarchy of dynamical timescales, does not by itself guarantee the appearance of a prethermal plateau. Under the same Hamiltonian, some observables may exhibit prethermal behavior, while others relax directly toward equilibrium. We uncover the mechanism governing this selectivity, showing that the emergence of a prethermal plateau depends not only on the Hamiltonian, but also on the observable and the initial state. Our results apply broadly to systems close to a fully permutation-symmetric limit and are illustrated with a long-range interacting spin model. We further prove that the Loschmidt echo provides a lower bound on the prethermal lifetime of any bounded observable whenever the prethermal and exact dynamics are unitary.
Read the full abstract

Quantum Physics

Depth-1 expanders on the unitary group and applications

We construct a constant-degree and constant-gap quantum expander on $n$ qubits where each unitary can be implemented by a depth-$1$ and 1D circuit of Pauli or CNOT gates. We provide two applications of this expander. First, we use it to construct a family of frustration-free 1D Hamiltonians whose ground states obey the entanglement-gap relation $S = \Theta(\Delta^{-1/2})$; this is believed to be optimal, but achieving it had been open. Second, we use it to provide a streaming protocol that tests for closeness to a class of 1D volume-law entangled states. Moreover, we extend our quantum expander to a constant-degree and constant-gap expander on the unitary group where each unitary is a single $T$ gate, a single $T^{\dagger}$ gate, or a depth-$1$ Clifford circuit. This implies that a random sequence of unitaries from the expander yields a gapped walk on a dense subgroup of the unitary group. This improves upon previous work by Bourgain and Gamburd which did not control the dependence of the gap on the dimension.
Read the full abstract

All authors Anurag Anshu, Shankar Balasubramanian, Jonas Haferkamp, Aram W. Harrow, Xinyu Tan

Computation and Language

Beyond Scores: Understanding LLM-as-a-Judge Mechanisms in Summarization Evaluation

LLM-based evaluators of natural language generation (NLG) quality are widely deployed as scoring tools and as automated training signals, yet the internal procedure by which they assign a rating remains poorly understood. We investigate this procedure mechanistically through an eight-attack perturbation taxonomy across the Readability and Adequacy dimensions of NLG quality, a generation pipeline that produces paired clean and corrupt summaries with controlled error intensity and explicit token-level modification maps, and a four-experiment battery of causal tracing, logit-lens vocabulary projection, and attention-head knockout applied to Themis (Llama-3-8B) and Prometheus (Mistral-7B). Both evaluators implement a structured, coherent evaluation pipeline operating in two stages: below layer 15, attention performs local error comparison and routes the result to the final input position; above it, the MLP cascade integrates the signal and writes the rating, with the decision crystallizing in the residual stream at a sharp late layer (L = 26 on Themis, L = 25 on Prometheus). Furthermore, a base-model control at the same scale (Llama-3-8B) reproduces the routing architecture and crystallization but not the stage separation, isolating the two mechanisms that fine-tuning specifically installs, suppression of below-L15 MLP contribution at the last position and a two-layer advance of the crystallization depth, indicating that fine-tuning sculpts an existing substrate rather than building the pipeline from scratch. We release the source code and data at https://github.com/himil-v/judge-mech
Read the full abstract

Software Engineering

Efficient SWE Agent Benchmarking via Trajectory-Aware Evaluation

Evaluating software engineering agents on realistic benchmarks is costly, since each task may require multi-step code exploration, modification, and test execution. Existing efficient evaluation methods select representative subsets to estimate full-benchmark performance, but are largely result-only: they fit historical pass/fail response matrices or static task semantics, discarding how agents solve problems. We propose PTA-IRT, a Privileged Trajectory-Aware Item Response Theory framework that fuses process and outcome signals. Historical execution trajectories supply process-level evidence beyond pass/fail, such as explored context, attempted edits, and solving paths, which PTA-IRT uses as privileged information for calibration subset selection and ability estimation. Under low calibration budgets, PTA-IRT consistently outperforms prior IRT baselines on score and ranking recovery across four SWE benchmarks. Code and data are publicly available at https://github.com/DeepSoftwareAnalytics/PTA-IRT.
Read the full abstract

All authors Kefeng Duan, Dewu Zheng, Yanlin Wang, Xiwen Wang, Ensheng Shi, Xilin Liu, Yuchi Ma, Jiachi Chen, Mingwei Liu, Zibin Zheng

Superconductivity

Singular Weak-Field Thermodynamics of 2D Superconductors

In a bulk 3D type-II superconductor, the lower critical field at which an isolated vortex becomes thermodynamically favorable is a size-independent material property. We show that the situation is different in 2D superconductors: the larger the superconductor, the weaker the field needed to create its first vortex. The lower critical field in 2D is always size-dependent. For a disk of area $\mathcal A$, the lower critical field $B_v(\mathcal A)$ scales as $\mathcal A^{-1}\ln(\mathcal A/\mathcal A_0)$ in the weak-screening regime and as $\mathcal A^{-1/2}$ in the strong-screening regime. We derive these results from an analytically tractable microscopic model that admits many-body wavefunctions for both the uniform and singly quantized vortex states in a magnetic field, and incorporate screening by coupling their long-distance 2D supercurrents to 3D Maxwell equations. These results motivate organizing the weak-field ground-state of a 2D superconductor in the $(1/\mathcal A,B)$ plane. The origin represents the zero-field thermodynamic limit and it is singular. Approaching the origin along the $B$ axis leads to an increasingly dilute vortex lattice, whereas approaching along the $1/\mathcal A$ axis yields the uniform vortex-free state. Our theory shows that every trajectory carrying fixed finite flux ultimately approaches the vortex-free state in the thermodynamic limit and provides a firm microscopic foundation for the weak-field thermodynamics of 2D superconductors.
Read the full abstract

Software Engineering

Adaptive Critical Token-Aware Retrieval for Repository-Level Code Generation

The repository-level code generation task requires synthesizing code that satisfies task requirements while remaining consistent with the target repository context. Since real-world repositories often exceed the input length limits of LLMs, existing approaches commonly adopt retrieval-augmented generation (RAG) to provide repository-specific context. Despite improving repository-context retrieval, existing methods typically provide context as task-level support, without explicitly identifying the critical tokens that require fine-grained repository context during generation. During the autoregressive generation process of LLMs, errors often concentrate at a small number of decisive positions: once such tokens are generated incorrectly, subsequent code may follow an incorrect semantic path and eventually lead to functional failure. We refer to these positions as "critical tokens". In this paper, we propose ACToR, an adaptive critical token-aware retrieval framework for repository-level code generation. ACToR identifies critical tokens during generation and triggers targeted retrieval on demand to provide repository context at these decisive positions. In addition, we design a position-aware weighting method for dense retrievers to prioritize context that is more informative for generation. We evaluate ACToR on two representative repository-level benchmarks, RepoExec and CoderEval. Experimental results show that ACToR consistently outperforms state-of-the-art methods, achieving relative improvements of 8.4% on RepoExec and 15.4% on CoderEval. Beyond performance gains, we systematically quantify the impact of critical tokens, revealing their central role in major generation failures and highlighting the necessity of targeted retrieval strategies. We provide the code and data at https://github.com/DeepSoftwareAnalytics/ACToR.
Read the full abstract

All authors Kefeng Duan, Dewu Zheng, Yanlin Wang, Terry Yue Zhuo, Mingwei Liu, Jianxing Yu, Jiachi Chen, Ensheng Shi, Xilin Liu, Yuchi Ma, Zibin Zheng

Computation and Language

CordisBench: Can Language Models Reason About Component Lifecycles in Dynamic Agent Harnesses?

Dynamic agent harnesses let language models change the software that shapes their own execution. This flexibility brings a new reasoning burden: a local plugin change can propagate through dependencies and cleanup. We introduce CordisBench, a 1,200-question benchmark of this lifecycle reasoning. It combines a controlled formal setting with programs executed against Cordis, a runtime that manages component dependencies and cleanup, and asks models to identify affected components, predict state after a specified teardown order, determine which conditions hold under all or some orders, and choose reconfigurations that succeed when executed. Across these tasks, we evaluate three efficiency-oriented models at low reasoning effort with 2, 4, 8, 16, 24, or 32 relevant interactions, using deterministic task-specific scoring. Models usually handle small systems well but grow less reliable as more interactions become relevant, especially when predicting final state and when reasoning across teardown orders. Additional inference effort recovers marked gains for some models. The cost is nontrivial: on our 16-interaction subset, GPT-5.6 Luna uses nearly 3,000 reasoning tokens per question at medium effort. For these controlled instances, that cost is avoidable: an independent finite reference semantics agrees with Cordis execution on every observation and action outcome used for scoring across all 528 executable questions.
Read the full abstract

Earth and Planetary Astrophysics

Constraining the Planetary Obliquity Distribution of Warm Jupiters

Warm Jupiters are an intriguing class of planets with uncertain origins. Their planetary obliquities could help distinguish between different formation pathways: planet-planet scattering and migration across resonances can excite large obliquities, whereas in-situ formation would more naturally produce low obliquities. We searched for oblateness-related anomalies in the transit light curves of six observationally favorable warm Jupiters: TOI-201b, TOI-1670c, TOI-199b, Kepler-9c, Kepler-30c, and Kepler-553c. Each planet's light curve is consistent with a spherical planet and provides degenerate constraints on the planet's sky-projected oblateness and obliquity. To overcome these limitations, we performed hierarchical Bayesian modeling of the population-level obliquity distribution. Assuming warm Jupiters are as oblate as Saturn ($f \approx 0.1$), we find their median obliquity to be below $12^\circ$ with $90$% confidence and below Saturn's obliquity ($27^\circ$) with $93$% confidence. Jupiter-like oblateness ($f \approx 0.06$) and larger obliquities are allowed. Simulations of JWST observations predict that significantly tighter constraints can be derived.
Read the full abstract

Computer Vision

UI-VISA: U-Net Initialized Vascular Image Segmentation Architecture

Accurate segmentation of vascular structures in digital subtraction angiography (DSA) images remains challenging due to the thin, elongated, and branching nature of blood vessels. Pixel-wise deep learning approaches such as U-Net achieve strong general-purpose segmentation performance but often produce fragmented or discontinuous predictions in fine vascular regions, since they do not explicitly enforce structural connectivity. Region growing algorithms preserve spatial context and topological continuity, but are highly sensitive to seed point initialization and can be computationally expensive. We propose UI-VISA (U-Net Initialized Vascular Image Segmentation Architecture), a hybrid pipeline that combines the complementary strengths of both approaches. UI-VISA uses U-Net's foreground predictions as informed seed points for a CNN-guided region growing algorithm, which then iteratively refines the segmentation by enforcing local connectivity and recovering fine vessel details that U-Net alone tends to miss or over-predict. We evaluate UI-VISA against standalone U-Net and a prior region-growing-based method (VISA) using 5-fold cross-validation on 26 DSA images. UI-VISA achieves the highest mean Dice and clDice scores across folds, and a paired Wilcoxon signed-rank test shows the improvement in clDice is statistically significant ($p=0.023$), consistent with the method's design goal of preserving vascular connectivity, while the improvement in Dice does not reach significance ($p=0.104$).
Read the full abstract

Computation and Language

The Rise of Verbal Reinforcement Learning

Natural language is emerging as a primary feedback channel for improving language agents, capable of conveying intent, preferences, and causal structure in forms interpretable by both humans and modern language models. We call this paradigm Verbal Reinforcement Learning (VRL) and offer the first unified account of it. We organize the field around a single axis, when verbal feedback takes effect in an agent's lifecycle and what it modifies, yielding three pillars: (1) Language as Grounding Signal, where language defines the task itself by specifying goals, states, and reward structures; (2) Language as Deliberative Feedback, where natural language guides reasoning at test time without the need to update model parameters; (3) Language as Learning Signal, where language-based feedback shapes model parameters through training. Within each pillar, we synthesize representative work, distinguish key subcategories of approaches, and outline the distinct role language plays in shaping agent behavior. Together, this taxonomy shows how verbal reinforcement is reshaping agent development, while also defining the challenges and opportunities for building more capable and aligned agents.
Read the full abstract

All authors Kshitij Tayal, Arun Sharma, Genta Indra Winata, Anirban Das, Sambit Sahu

Robotics

Facet-0: A Robotic Foundation Model for Contact-Rich Precise Manipulation

Real-world robotic assembly at sub-millimeter tolerances demands spatial precision, compliant interaction, and robustness to contact failures. We present Facet-0, a robotic foundation model that predicts and values the contact consequences of its actions. Facet-0 unifies multimodal representation learning and reinforcement learning (RL) post-training around a joint action-wrench proposal: a causal wrench history is aligned with vision-language semantics and kinematic state, and flow matching generates each action chunk together with the future wrist-wrench profile it is expected to induce. Deployment rollouts train a distributional Action-Wrench Critic to distinguish motions with similar task progress but different contact outcomes, while phase-aware rewards and contact-selective credit concentrate policy improvement on decisive interactions. To accommodate part-specific dynamics, a lightweight bounded actor reuses the frozen representation for on-robot adaptation; RL remains defined over executable Cartesian actions, while an auxiliary wrench head preserves predictive, non-commanded action-contact coupling. Trained on ManuFacet-1K, a 1,000-hour force-synchronized corpus spanning three embodiments and multiple manufacturing cells, the bounded task-adapted system reaches 82% mean success on five sub-millimeter computer-assembly tasks, compared with 15% for the strongest baseline, with 0.5 mm placement accuracy and 50 ms command latency.
Read the full abstract

All authors Haoyuan Deng, Haichao Liu, Wenkai Guo, Yuan Ling, Zaijia Yang, Yuanjiang Xue, Haosheng Sun, Liangzi Wang, Ziwei Wang

Theoretical Economics

Mechanism Design for Alignment and Control

We develop a framework for mechanism design with AI agents whose alignment (preferences) and capabilities (feasible actions and information) are unknown. We want such agents to act on our behalf so mechanisms must incentivize both honesty and obedience. A one-sided imitation structure—capabilities can be concealed but not counterfeited—yields a revelation principle, a characterization of implementable policies via nested cyclical monotonicity, and conditions under which eliciting higher-order beliefs can discipline multiple agents. We apply our framework to stylized examples of (i) sandbagging in which a more capable agent pretends to be less capable; (ii) an alignment–interpretability trade-off, where the two are substitutes in the instrument but complements in value; (iii) discipline via peer scoring; (iv) coupling rewards to induce competition among multiple agents; and (v) scalable oversight and reward shaping.
Read the full abstract

Number Theory

Ten unknowns for Hilbert's tenth problem over the integers

Hilbert's Tenth Problem over the integers was solved negatively by Y. Matiyasevich in 1970. In this paper, we prove that there is no algorithm to determine for any polynomial equation $P(z_1,\ldots,z_{10})=0$ (with integer coefficients and ten unknowns) whether it has integer solutions. This improves the previous 11 unknowns theorem.
Read the full abstract

Superconductivity

Understanding the superconducting proximity effect in semiconductors through quantum oscillations

Superconductor-semiconductor hybrids host emergent states of matter and offer a platform for new qubits, but the superconducting metal shunts electrical transport, which rules out conventional semiconductor characterization and leaves the hybrid parameters to speculation. Here we determine density, mass, $g$-factor, mobility and subband occupation beneath the superconductor, from Shubnikov-de Haas oscillations of a buried InAs quantum well under Al, Sn, V, Nb, Ta and Re films, with a Dingle analysis that accounts for the shunt. Every metal adds an interface subband whose occupation falls into one of two classes, whereas the mass and $g$-factor of the buried well are unchanged to within 10%. Within the uncertainty set by the transport mobility, no film shortens the quantum lifetime of the buried well, and Al and Sn lengthen it. Quantum lifetimes bound the hybridization of the interface subband to 2-4~meV. These measurements supply the normal-state parameters that tunnelling spectroscopy renormalizes but cannot measure.
Read the full abstract

All authors Milo Coombs, Teun A. J. van Schijndel, Yu Wu, Jason T. Dong, Yilmaz Gul, Julian Choi, Christopher J. Palmstrøm, Greg P. Mazur

High Energy Physics (Phenomenology)

Fermionic Dark Matter Absorption and the High-Energy Event in LUX-ZEPLIN

The LUX-ZEPLIN (LZ) experiment has reported a single candidate event in the high-energy nuclear recoil window $248\pm32.5\ \mathrm{keV}_{\mathrm{nr}}$ with an exposure of $2.84\ \mathrm{ton}\cdot\mathrm{yr}$, while the low-energy spectrum remains consistent with background expectations. We demonstrate that this excess can be naturally explained by the neutral-current absorption of fermionic dark matter on xenon nuclei. For a dark matter mass $m_\chi \simeq 247\ \mathrm{MeV}$, the coherent absorption process produces a monoenergetic nuclear recoil at $E_R \simeq 248\ \mathrm{keV}_{\mathrm{nr}}$. At this momentum transfer, the absorption process enters the incoherent regime, where scattering off individual nucleons produces a broad recoil spectrum extending from $\sim 200\ \mathrm{keV}$ to $100.2\ \mathrm{MeV}$. We show that a single effective field theory coupling can simultaneously produce one event in the $248\pm32.5\ \mathrm{keV}_{\mathrm{nr}}$ window while remaining consistent with the non-observation of events in neighboring energy regions. The required single-nucleon absorption cross section is $\sigma_{\chi N}^{\mathrm{NC}} = 1.07\times10^{-46}\ \mathrm{cm}^2$, corresponding to an effective field theory scale $\Lambda \simeq 11.5\ \mathrm{TeV}$. However, a recasting analysis of KamLAND data on the neutron-emission channel $\chi+{}^{12}\mathrm{C} \to \nu + n + {}^{11}\mathrm{C}^*$ excludes this benchmark parameter space, establishing a significant tension between the LZ excess interpretation and existing constraints from large-volume scintillator detectors. We discuss the implications of this tension and prospects for resolving it with future dedicated high-energy analyses.
Read the full abstract

Computation and Language

StudentSim: Training LLM-based Student Simulators

AI tutors are most useful when they adapt to each student's strengths, weaknesses, and preferred guidance, but evidence about which guidance works for which student is sparse, slow, and costly to collect from real learners. Student simulators can provide this signal as a proxy, yet existing approaches are limited: state-tracking models fit student behavior but struggle to process explanations or corrections, while LLM role-play follows guidance fluently but does not reliably match the competence of the student being imitated. We present StudentSim, a training framework that turns sparse per-student data into individualized simulators through pooled training followed by per-student specialization. The resulting simulators both mirror a student's own responses and update them under tutor guidance. We also introduce StudentSimEval, a standardized protocol covering 60 students across chess, second-language English writing, and mathematics, using public learner datasets with de-identified records shared for research. StudentSimEval measures behavioral fidelity (F), or how well a simulator matches a student's responses, and guidance responsiveness (R), or how readily it updates under tutor guidance, with all methods fit and evaluated on the same records. Across all three domains, StudentSim outperforms GPT-5.4 on both metrics. In chess, StudentSim reaches F=0.51 and R=0.91, compared with 0.23 and 0.72 for GPT-5.4 and 0.45 and 0.27 for Maia2. As a proof of concept, using StudentSim as a reward model for tutor reinforcement learning produces a chess tutor that expert humans rate as more accurate, better-guided, and more personalized than a no-RL baseline and a tutor trained against a GPT-5.4 simulator reward. Code is available at https://github.com/microsoft/StudentSim.
Read the full abstract

All authors Ke Yang, Chenglong Wang, Michel Galley, Chandan Singh, Jeevana Priya Inala, ChengXiang Zhai, Jianfeng Gao

High Energy Physics (Phenomenology)

TeV Higgsino Dark Matter from LZ Nuclear Recoil to Fermi-LAT Gamma Rays

The excess of high-energy nuclear recoils, which was recently reported by the LUX-ZEPLIN (LZ) experiment, together with the Fermi-LAT Galactic-Center preference, suggests new physics beyond the standard model (SM). One of the minimal extensions of the SM is to introduce a fermionic electroweak doublet in which the lightest neutral component can play the role of a stable dark matter particle. We show that the minimal higgsino dark matter can offer a common origin for both excesses. The inferred mass splitting, however, depends on the local dark matter velocity distribution, and including the high-velocity tail generated by the gravitational influence of the Large Magellanic Cloud yields $\delta_m =481$–$519~\mathrm{keV}$ for $m_\chi =1000$–$1183~\mathrm{GeV}$ at $1\sigma$ level. Near elastic-scattering blind spots, solar-captured higgsinos fail to reach thermal equilibrium, weakening their annihilation signal, so that ten years of IceCube data exclude $\delta_m \lesssim 475~\mathrm{keV}$ at 90% C.L. The viable parameter space can be tested through the associated gamma-ray line, which is within the projected reach of CTAO.
Read the full abstract

Analysis of PDEs

Smooth expanding planar simple waves for Euler–Poisson–Boltzmann: uniform stability and quasineutral expansion

We study the warm-ion Euler–Poisson system with Maxwell–Boltzmann electrons on the cylinder $\R\times\T$ in the quasineutral regime. Taking a smooth expanding planar simple wave of the effective Euler system as the reference state, we construct an even Debye expansion through arbitrary finite order $M$ with a residual of $O(\eps^{2M+2})$. We rigorously establish nonlinear stability on every fixed interval $[t_0,T]$ ($t_0\ge0$), achieving a lifespan and energy constants strictly independent of the Debye length $0<\eps\le\eps_0$. To overcome the singular scaling of the electric field, we develop a novel compensated energy topology that couples the warm-ion symmetrizer to the time-differentiated nonlinear Poisson constraint. By integrating the top-order electric work directly into the time derivative of a weighted energy functional, this mechanism controls the potential in $H^s$ and its gradient in $\eps H^s$, completely eliminating the $\eps^{-1}$ loss typically encountered in the momentum equation. This framework successfully governs distinct neutral end states, captures genuinely two-dimensional rotational perturbations, and yields an arbitrary-order quasineutral asymptotic expansion for prepared data, providing a critical analytical foundation for the geometric theory of multidimensional quasineutral rarefactions.
Read the full abstract

Human-Computer Interaction

Designing Proactive Thought Partners for Writing

Writing involves diverse cognitive activities, from ideation to revision, and writers' needs vary across individuals and moments. Proactive AI promises to provide the right support at the right time, yet existing proactive tools largely focus on generic textual assistance, such as autocomplete. This paper studies the design space of proactive thought partners: AI agents that proactively offer customizable, higher-level cognitive support during writing. We instantiated this concept in a technology probe and deployed it with 16 participants for one week. The probe allows users to create partners by configuring their roles and proactivity. As users write, relevant partners take the initiative at appropriate moments to offer suggestions. Our findings show that participants configured proactive support through prospective planning, used suggestions for both idea generation and self-monitoring, and valued lightweight visual representations alongside non-directive rhetorical framing for non-intrusive interventions. We derive implications for designing proactive writing assistants around customization, timing, engagement, and representation.
Read the full abstract

All authors Chao Zhang, Abe Davis, Chih-Wei Chen, Chin-Chia Hsu

Machine Learning

The Structure of Quantization Damage in LLMs: Why the Next Bit Should Be Spent Globally

Post-training quantization (PTQ) is widely used to reduce the cost of serving large language models (LLMs), but its accuracy cost is uneven and is often tuned per model. We study where quantization damage occurs and how to allocate a small additional precision budget. Using causal mixed-precision intervention as ground truth (raise each layer to 8-bit in turn and measure the accuracy it recovers) across 9 open-weight models in 4 architecture families, we test 3 intuitive hypotheses: that quantization damage lives in task circuits, where the model computes, or in weight statistics. None of them predicts which layers benefit from restored precision. Recovery is instead diffuse: for 8 of 9 models, recovering 75% of the gap takes roughly half the layers; the lone exception, Qwen3-8B, is sharply concentrated. At a matched precision budget, spending it globally on finer quantization granularity beats locally repairing the most recoverable layers for all 8 group-128-compatible models (all but OpenLLaMA, whose width rules out group-128), by 21-52 points, including the concentrated Qwen3-8B. We report 2 secondary findings: the residual is budget-limited (8-bit is near-lossless in our evaluation across RTN, GPTQ, and AWQ), and the location of peak recovery correlates with architecture within a family, though not across families. Within this budget setting, global granularity is a better default than selectively protecting critical layers. More broadly, cheap signals that correlate with quantization damage do not necessarily identify where restoring precision improves accuracy; this must be tested with causal intervention.
Read the full abstract

Astrophysics of Galaxies

The JWST Emission Line Survey (JELS): A narrow-band determination of the H$\alpha$ Luminosity Function and Cosmic Star Formation into the Epoch of Reionization

The recent star-formation activity in galaxies can be optimally traced by the H$\alpha$ emission line, with the resulting H$\alpha$ luminosity function (LF) at a given epoch providing a reliable probe of cosmic star formation. We present the first narrow-band determined H$\alpha$ LF into the Epoch of Reionization (EoR) at $z\sim6.1$, using 39 H$\alpha$ emitters selected from the JWST Emission Line Survey (JELS). The observed and dust-corrected LFs are broadly consistent with recent slitless spectroscopic measurements but show notable discrepancies with predictions from cosmological simulations, likely reflecting differences in emission-line and dust modelling. Fits combining multiple LF datasets help constrain the high-redshift faint-end slope of the H$\alpha$ LF ($-1.79<\alpha_{\rm{H\alpha}}<-1.62$), but there remains uncertainty in its evolution with redshift. Integrating the JELS dust-corrected H$\alpha$ LFs yields a star-formation rate density of $\log_{10}(\rho_{\rm{SFR_{H\alpha}}}\,/\,\rm{M_{\odot}\,yr^{-1}\,Mpc^{-3}})=-1.93\,^{+0.14}_{-0.12}$ or $-2.00\,^{+0.16}_{-0.10}$, assuming a continuum-to-line extinction ratios $\eta_{\rm{dust}}=A_{\rm{cont}}(\rm{6563\,\mathring{A}})/A_{\rm{H\alpha}}=1$ and 0.44, respectively. Both measurements are consistent within uncertainties with previous results using standard assumptions for the LF integration limit ($L_{\rm{H\alpha,\,lim}}$) and SFR calibration constant $\kappa_{\rm{H\alpha}}$, despite the uncertainties in the dust corrections. We explore the metallicity dependence of $\kappa_{\rm{H\alpha}}$ and find $\rho_{\rm{SFR_{H\alpha}}}$ decreases 0.43 dex compared to the fiducial result and is no longer consistent with UV-determined $\rho_{\rm{SFR}}$ at $z\sim6$. This work highlights the importance of narrow-band surveys in probing the faint H$\alpha$ population and providing new constraints on cosmic star-formation activity into the EoR.
Read the full abstract

All authors C. A. Pirie, K. J. Duncan, P. N. Best, D. J. McLeod, C. L. Hale, R. K. Cochrane, S. R. Flury, H. M. O. Stephenson, M. I. Arnaudova, M. Brinch, J. S. Dunlop, E. Ibar, Z Li, J. Matthee, R. J. McLure, L. Ossa-Fuentes, A. L. Patrick, J. Selfridge, Ian Smail, D. Sobral, J. P. Stott, A. M. Swinbank

Dynamical Systems

Families of relative periodic orbits in the planar three-body problem via consecutive alignments

Relative periodic orbits (RPOs) are solutions of the three-body problem that are periodic in a uniformly rotating reference frame and, in general, quasi-periodic in inertial coordinates. We present a numerical procedure for computing and continuing one-parameter families of RPOs of the planar Newtonian three-body problem. The method exploits consecutive syzygies, understood here as configurations in which the three bodies are aligned and their velocities satisfy the corresponding symmetry conditions. Matching the positions and momenta at two consecutive alignments reduces the computation of RPOs to a low-dimensional nonlinear problem. Its solutions are then numerically continued, and linear stability is determined from the nontrivial eigenvalues of the rotated monodromy matrix after removing the neutral directions associated with conserved quantities and continuous symmetries. The procedure is applied to several mass distributions and initial configurations, producing families of Poincaré, Hill, and binary-type solutions. These families exhibit transitions from nearly circular to highly eccentric motion, changes of stability near resonances and turning points, and absolute periodic solutions when the rotation angle is a rational multiple of 2{\pi}. In the Hill families, the continuation connects satellite configurations with circumstellar motion as the smallest body loses its gravitational binding to the intermediate body. Circumbinary and circumstellar configurations are also obtained in the binary regime. The results illustrate the dynamical diversity of RPOs and provide coherent three-body motions that can be used as prescribed trajectories in restricted four-body models.
Read the full abstract

Computer Vision

A Benchmark for Vehicle Attribute Classification in Cross-Domain Surveillance Scenarios

Vehicle attribute analysis is a key component of Intelligent Transportation Systems (ITS), supporting applications such as vehicle identification, traffic monitoring, and forensic investigation. However, models trained under controlled conditions often degrade in real surveillance scenarios due to changes in viewpoint, occlusion, illumination, and sensor characteristics. This paper introduces Unconstrained Vehicle Identification Benchmark (UVIB), a benchmark for evaluating three operational vehicle-analysis tasks: front/rear orientation, occlusion-related suitability for Vehicle Make and Model Recognition (VMMR), and color clarity. The benchmark contains 84,835 vehicle images from seven public Brazilian datasets, grouped into surveillance and general acquisition domains, with unified binary annotations that were not jointly available in the original sources. Four representative architectures, EfficientNetV2-S, ResNet-50, ViT/B-16, and YOLO11s-cls, are evaluated under mixed-domain, cross-domain, and cross-dataset protocols. The results show that domain shift has a stronger impact than architecture choice, with substantial degradation in cross-domain settings, especially for VMMR suitability and color clarity. While orientation generalizes more reliably, VMMR suitability remains affected by class imbalance and ambiguous occlusions, and color clarity is highly sensitive to illumination and sensor modality. These findings highlight the need for benchmarks and evaluation protocols that explicitly measure operational robustness beyond standard in-domain accuracy. The proposed benchmark is publicly available at https://github.com/UFPR-IPASP-PR/uvib-vehicle-attributes/.
Read the full abstract

All authors Sergio M. Silva Jr., Otavio T. Remer, Gabriel E. Lima, Lucas Wojcik, Rayson Laroca, David Menotti