Materials Science
Papers
Fresh from arXiv, typeset properly. Highlighted words trade on the market.
All of arXiv
Thursday 10 September
Ultrafast photoinduced insulator-to-metal transitions in correlated materials are often mediated by lattice distortions, yet the role of interfacial lattice constraints in shaping nonequilibrium pathways remains largely unexplored. We use azimuth-resolved time-resolved X-ray diffraction to track orientation-dependent strain dynamics in granular V$_2$O$_3$ thin films on c-cut sapphire, where thermal-expansion mismatch imposes anisotropic interfacial strain. Across the thermal transition, the azimuthal profile of the (110)$_H$ strain inverts curvature, providing direct evidence of partial clamping of the hexagonal basal-plane lattice (a$_H$,b$_H$). After photoexcitation of the antiferromagnetic insulating phase, the structural response remains clamp-limited: weakly constrained grain families reach the full basal-plane contraction characteristic of the metallic-like state, whereas strongly constrained families exhibit a strongly reduced distortion. Fluence-dependent measurements further disentangle transformed fraction from clamping-limited lattice distortion. Our results show that interfacial clamping acts as a static selector for ultrafast phase switching and provide a general route to quantify anisotropic strain dynamics in heterostructures.
Information Theory
Near-Gaussian counterexamples to the Ball-Nayar-Tkocz entropy concavity conjecture
The Ball-Nayar-Tkocz conjecture asserts that, for independent real random variables $X,Y$ with a common log-concave density, the function $t\mapsto h(\sqrt tX+\sqrt{1-t}Y)$ is concave on $[0,1]$. Gaussian distributions have this property. We construct strongly log-concave counterexamples arbitrarily close to the Gaussian and study their persistence under Gaussian smoothing. Let $X$ have mean zero and variance one, let $G$ be an independent standard Gaussian, and let $q\in[0,1]$ denote the signal variance proportion in the smoothed variable $\sqrt qX+\sqrt{1-q}G$. In the asymmetric case, for every fixed $0<q\le1$, there are counterexamples for which the entropy of the weighted sum of two independent copies of the smoothed variable has strictly positive second derivative near the endpoints. In the symmetric case, for every $4/7<q\le1$, there are counterexamples for which an unequally weighted sum has strictly greater entropy than the equally weighted sum. Both families have smooth, strictly positive densities and are strongly log-concave before and after smoothing. The density ratios $f/\varphi$ converge uniformly to one, and each derivative of fixed positive order converges uniformly to zero, where $\varphi$ is the standard Gaussian density. The symmetric counterexamples can also match any prescribed finite number of Gaussian moments exactly. The asymmetric construction uses an endpoint expansion of entropy, whereas the symmetric construction exploits the different decay rates of high-order Hermite perturbations under different weights. Thus, densities violating concavity approach the Gaussian density in the strong sense described above. The question of uniform concavity in the symmetric class for $0<q\le4/7$ remains open.
Read the full abstract
High Energy Physics (Theory)
Gauss Law, Monodromy Defect, Magnetic Lattice Translation, and Lattice Duality
In a Hamiltonian lattice gauge theory, the physical Hilbert space is fixed by the Gauss law, by choosing an eigenspace of the Gauss law operator which generates the gauge transformation. Choosing a non-trivial eigenspace is referred to as the modified Gauss law. In 2+1d dynamical U(1) pure gauge theory, we show that these physical Hilbert spaces are defect Hilbert spaces associated with non-topological monodromy defects for monopole operators and topological defects for U(1) 1-form symmetry, inserted along the time direction. We also construct the corresponding lattice translation operators. Under a nontrivially modified Gauss law, these translations obey a magnetic translation algebra. Finally, we discuss how the modified Gauss laws are manifested in 2+1d compact bosons, via exact duality on the lattice.
Read the full abstract
Quantum Physics
The Quantum Plumber's Problem
Recent work quantified the notion of quantum counterfactual gain for an extended Elitzur-Vaidman bomb test style scenario, through a connection to the negativity of the Kirkwood-Dirac quasiprobability distribution. We here extend this work to identifying quantum advantage in a new scenario, which we term the “Quantum Plumber's Problem”. In this scenario, we imagine a “quantum plumber”, who knows that one path of an interferometer is blocked, and wants to find the optimal strategy for identifying with certainty which path this is. We discuss various strategies for a generalised path-encoded interferometer, as well as for the specific case of Hofmann's three-path interferometer, introduced in a recent analysis of the relationship between states in five measurement contexts of a three level system. We support our arguments on the relative merit of competing strategies with data collected over many simulated attempts at locating blockages. We also present results for a variant of the game in which the blockage is replaced by a non-demolition detector.
Read the full abstract
Statistical Mechanics
Rare-History Transitions in Temporally Random Integrable Quantum Circuits
We study current fluctuations in a temporally random integrable quantum circuit. Commutativity reduces every drive history exactly to its layer composition, turning annealed fluctuations into a competition between current gain and the large-deviation cost of rare compositions. For each fixed composition, homogeneous thermodynamic Bethe ansatz dressing supplemented by ballistic fluctuation theory yields the conditional current statistics. Their annealed large-deviation contraction predicts a first-order switch between two dominant history classes, terminating on a line of regular cusp endpoints. Finite-time analysis shows how the switch is rounded. Thus temporal randomness can act as an emergent order-parameter-like coordinate in trajectory space.
Read the full abstract
Computation and Language
SWRouter: Similarity-Contractive Window Routing for Multi-Turn Large Language Model Conversations
Large language models exhibit complementary strengths, motivating routing methods that dispatch each query to the most suitable model. Although existing routers are effective in single-turn settings, they do not directly transfer to multi-turn dialogue, where routing performance critically depends on how historical context is segmented, retained, and incorporated into the current prompt. This introduces two fundamental challenges: preventing information loss and information confusion during context construction, and evaluating routing quality without conflating model selection with prompt construction quality. In this paper, we propose SWRouter, a Similarity-Contractive Window Router for multi-turn large language model routing. SWRouter combines a similarity-based context segmentation mechanism for prompt construction with a dual-metric evaluation framework that decouples construction accuracy from router performance. Experiments on multi-turn dialogue benchmarks demonstrate that SWRouter consistently surpasses strong baselines, achieving a 16.26% improvement in evaluation accuracy over the best individual large language model and an additional 8.22% gain over the Conv-ID Context baseline. Our results highlight that multi-turn large language model routing requires a joint design of context construction and evaluation, rather than a direct extension of single-turn routing methods.
Read the full abstract
Cryptography and Security
Heterogeneous Cross-Chain Transaction Tracing for Solana Bridges via Candidate-Set Selective Decision
Solana is a rapidly growing high-throughput blockchain platform that has attracted substantial liquidity and user activity. However, this expansion has also drawn the attention of illicit actors, who frequently leverage cross-chain bridges to route illicit funds onto Solana to obfuscate transaction lineage. Unlike EVM-compatible platforms, Solana features distinct execution dynamics and lacks standard event logs, creating severe semantic gaps that prevent existing tracing methods from reliably correlating cross-ledger transactions. In this paper, we formalize four types of Solana-bound cross-chain transaction modes and propose a candidate-set selective decision-based tracing method called SolTracer. SolTracer maps disparate execution semantics into a unified event space and employs candidate-set selective decision-making to reliably associate target transactions while abstaining when valid targets are absent. Extensive experiments demonstrate that SolTracer outperforms state-of-the-art (SOTA) methods across three representative scenarios: closed-world association, open-world association, and cross-source-chain generalization. In particular, under the challenging open-world setting with a 50% TA ratio, SolTracer improves the F1 score by 20.16% over the strongest SOTA baseline. Utilizing SolTracer, we conduct an empirical analysis on real-world cross-chain transfers to investigate ecosystem dynamics. Our analysis explores the stark count-value divergence across bridge mechanisms, the prevalence of cross-asset shifts, and the decoupling between on-chain settlement and explorer visibility.
Read the full abstract
Sound
X-AuT: Progressive Audio-Encoder Compression for Speech LLMs with Cross-Scale Distillation
Reducing audio-encoder depth lowers the inference cost of speech large language models, but removing complete blocks perturbs the embeddings consumed by the decoder and can cause deletion and premature end-of-sequence errors. We introduce X-AuT, a progressive framework that selects layer combinations through short behavioral probes and restores the pruned model through representation alignment, cross-scale distillation, scheduled student-policy supervision, and LoRA finetuning. The language-model backbone remains frozen, while attention LoRA adapters and the tied output embedding adapt during distillation. Training uses the highest-agreement tier from a transcript-consistency pipeline, followed by source reweighting during finetuning. On ten public Chinese–English benchmarks, compressing Qwen3-ASR-0.6B from 18 to 16 audio-encoder layers reduces macro-average error from 5.61% to 5.27%. The 14-layer model reaches 5.75% with 20.7% fewer audio-tower parameters. Under the matched recipe, the 1.7B teacher yields 5.55% mean error, compared with 8.45% for self-distillation, and progressive 18$\rightarrow$14 pruning outperforms direct pruning (5.75% vs. 6.73%). These single-run results establish two practical operating points and show that the accuracy effects vary across benchmarks. Project website: https://xpeng-ai.github.io/x-aut
Read the full abstract
Cryptography and Security
"They don't care about this": A Systematic Study of TEE Build Reproducibility in the Wild
Trusted Execution Environments (TEEs) have become a cornerstone of modern cloud computing, providing strong confidentiality and integrity guarantees for both code and data. A critical component of this trust model is remote attestation, which enables external entities to verify the authenticity and integrity of code executing within a TEE through cryptographic measurements. However, the effectiveness of remote attestation fundamentally depends on the verifier's ability to trace the reported measurement back to the original source code - a property that can only be guaranteed through reproducible builds. In this paper, we investigate the reproducibility of TEE builds through a technical analysis of 115 TEE deployments. Our analysis spans popular TEEs such as Intel SGX, Intel TDX, and AMD SEV, and reveals that a striking 91% of those deployments were not reproducible, with 80% failing to provide both source code and a reference build, the two essential prerequisites for reproducibility. To explore the root causes, we contacted the maintainers of 50 SGX projects and managed to recruit 12 developers from industry and academia for interviews. Only one of our participants reported that reproducibility is a priority during development, effectively confirming our technical findings. Beyond technical barriers (e.g., timestamps included in the binary) that can be readily addressed, we identify broader ecosystem-level challenges, such as the lack of control over the build environment in projects involving multiple stakeholders. We argue that achieving reproducibility in TEEs requires a holistic development approach that extends beyond individual developers and calls for stronger commitments - rather than treating TEEs as a "security badge".
Read the full abstract
Functional Analysis
Noncommutative commutant lifting, interpolation, and shift-invariant subspaces
We establish a Sarason-type commutant lifting theorem in the framework of noncommutative Drury-Arveson space and apply it to solve the noncommutative Nevanlinna-Pick interpolation problem on the noncommutative row ball. In particular, it recovers the classical Nevanlinna-Pick interpolation theorem over the open unit disc. We also provide a Beurling-Lax-Halmos-type characterization of shift-invariant subspaces of noncommutative Drury-Arveson space using dilation theoretic technique.
Read the full abstract
Combinatorics
Undirected edge geography games on stacked prism graphs
The undirected edge geography is a two-player combinatorial game on an undirected graph. The players start at the root vertex and alternately move the root along an incident edge to its other endpoint and then delete that edge. The first player who has no remaining move is the loser. For positive integers $m$ and $n$ where $m\geq 3$, the stacked prism graph is $SP(m,n)=C_m\square P_n$. In this paper, we completely determine the winner of the game on $SP(2m,n)$ for $m\geq 2$ and $SP(m,2)$ for $m\geq 3$, and provide a winning strategy for the winner.
Read the full abstract
Human-Computer Interaction
Are Caption Metrics Broken? Latency, Deaf and Hard of Hearing User Ratings, and Bias across Technologies
Live captions on TV often contain errors and timing issues, making it hard for deaf and hard-of-hearing (DHH) viewers to follow dialog. It is essential that caption quality metrics reflect the lived DHH TV viewing experience. To this end, we describe a U.S.-based large-scale online survey with 216 validated participants, who provided 302 responses containing a cumulative 4,832 data points. Participants viewed videos drawn from a pool of 70 clips recorded from live TV, and were asked to rate the caption quality and subjective understanding of the content across four conditions: TV captions as originally recorded with up to 7-12 seconds delay, TV captions synchronized with audio, Automatic Speech Recognition (ASR)-generated captions synchronized with audio, and ASR captions with an average two-second delay. All captions were evaluated against the Word Error Rate (WER), Automated Caption Evaluation (ACE2) and Number, Edition and Recognition (NER) metrics. Results show that TV and ASR captions were rated similarly. For TV captions, all three metrics were moderately-to-highly correlated with viewer ratings, but far less so for ASR captions, making them far from technology-neutral. Additionally, caption latencies significantly impact the viewer experience, especially typical 7-12-second TV delays. We discuss the implications for the adoption of caption quality metrics.
Read the full abstract
Commutative Algebra
Associated primes, witnesses, and omega invariants of monomial ideals
We introduce and study the omega invariant of a proper ideal in a Noetherian commutative ring, defined as the number of associated primes of the ideal. Our main objective is to investigate this invariant for monomial ideals and their powers. We characterize associated primes through monomial witnesses and provide an algorithmic procedure for constructing such witnesses from the exponent vectors of the minimal generators. These results lead to explicit formulas and bounds for the omega invariant without requiring the computation of a primary decomposition. We further establish alternative descriptions using irreducible decompositions and Alexander duality. A matrix-based approach is developed to detect associated primes of powers of monomial ideals directly from the exponent matrix of the original ideal. We also investigate the behavior of witnesses under passage from $I^n$ to $I^{n+1}$ and derive corresponding results for edge ideals of graphs.
Read the full abstract
Atomic Physics
Non-stick vacuum wall collisions with a laser-coolable molecule
Molecular species that are suitable for direct laser cooling are typically considered lost or destroyed if they collide with an ambient temperature vacuum wall. Here, we study surface collisions with aluminum monofluoride (AlF), a laser-coolable molecule that survives this process with unusually high probability. We detect the outgoing AlF molecules from a single wall collision via Doppler-sensitive laser-induced fluorescence spectroscopy, using incoming supersonic (pulsed) and thermochemical (continuous) molecular beams. The angular, velocity and rovibrational level distributions of the outgoing molecules show near-complete thermalisation to the wall in a single collision event. We determine an upper limit to the surface residence time of about 5$~\mu$s, and by monitoring the decay in density of pulses of molecules loaded into a small storage volume, we deduce the surface sticking probability for different materials. For a siloxane-coated metallic surface, the sticking probability of AlF is about 0.015, allowing us to accumulate molecules from the thermochemical source into an ambient temperature storage vessel at densities near $10^{8}~$cm$^{-3}$. This provides a route to compact, portable traps for neutral molecules.
Read the full abstract
Strongly Correlated Electrons
Instabilities in self-consistent diagrammatic approaches and how to cure them
While self-consistent diagrammatic approaches are widely used to compute the physical properties of correlated quantum materials, their applicability may get severely hindered precisely in the parameter regions where the most exciting physics is observed. One of the major issues, referred to as "misleading convergence", is the tendency of iterative schemes to converge to unphysical fixed points for intermediate-to-strong electronic interactions, regardless of the numerical accuracy of the computation. Here, we explicitly verify that the origin of this problem in several established self-consistent many-electron approaches, defined in the general diagrammatic framework of the boson-exchange formalism, resides exclusively in the stability condition of the respective iteration schemes, and not in an intrinsic breakdown of their self-consistent diagrammatic description. This insight enables a simple and general remedy, as recently proposed in Phys. Rev. Lett. 137, 016502 (2026): The redefinition of the iterative procedure, by inverting the unstable eigendirections of the Jacobian associated with the fixed point of the self-consistent algorithm. We illustrate the successful outcome of this procedure by means of systematic calculations performed on exactly solvable testbed models. Our results demonstrate that the physical fixed point of the diagrammatic schemes we considered can be stabilized, de facto, across the entire parameter range, including the most challenging nonperturbative/strong-coupling regimes.
Read the full abstract
Cryptography and Security
Deep-Fake CAPTCHA: Mitigating Next-Generation Social Engineering Attacks
This paper presents DF-CAPTCHA, an active defense against real-time deepfake impersonation in voice and video calls. Instead of passively searching for artifacts, DF-CAPTCHA prompts the caller to perform simple challenge-response tasks that are easy for humans but difficult for current real-time deepfake systems to generate convincingly. The framework verifies the response using four criteria: realism, identity consistency, task completion, and response time. We evaluate the approach across both audio and video modalities using user studies and experiments with real-time deepfake models. Results show that people often struggle to distinguish real-time deepfakes from authentic media, while DF-CAPTCHA substantially improves detection performance over passive methods, reaching high accuracy in both modalities. These findings suggest that active challenge-based verification is a practical and robust defense against next-generation social engineering attacks based on real-time deepfakes.
Read the full abstract
Artificial Intelligence
From Queries to Narratives: Cultural Heritage Data Stories for Knowledge Graph Exploration and Quality Assessment
Cultural-heritage KGs such as the NFDI4Culture-KG contain millions of triples about artworks, music, inscriptions, historical events, and the people and places connected to them. For many users, however, discovering this knowledge can be difficult. While SPARQL can be learned, writing meaningful queries first requires an in-depth understanding of the graph's data model, an investment many domain researchers and practitioners are unwilling to make. Even with existing user interfaces, a starting point and some guidance are usually needed, because the data contained in the graph is highly specialized, heterogeneous, and constantly growing, making it challenging to know what it contains or which questions it can answer. In this paper, we present data stories as a way not only to lower this barrier, but also to turn exploration into data-quality assessment, and thus combine accessible querying with the discovery of issues that remain hidden in aggregate statistics. In this contribution, a data story is understood as a narrative document that integrates explanatory text and images with executable SPARQL queries and their visualized results. It is described how they are authored against the graph and how they serve several purposes: guiding users through an unfamiliar graph, creating reproducible narratives, and surfacing data-quality issues previously hidden in aggregate statistics. The authoring platform LODEON including its Sparnatural and AI-supported authoring assistants is introduced as a proof-of-concept. Within the authoring environment, every claim made about the data can be backed by an explicit query, making these narratives transparent and reproducible. This paper also reflects on lessons learned from hands-on seminars and workshops. Early experience suggests that such data stories make cultural-heritage knowledge graphs more accessible for both exploration and quality assessment.
Read the full abstract
Quantum Physics
The Quantum Composition Paradox
Quantum theory does not generally permit the probability laws obtained from individual unitary steps by the Born rule to be sewn into a consistent genealogy; we classify the exceptions and show that faithful composition can hold from an initial boundary yet fail after an internal restart. For finite-dimensional, composition-closed unitary families, universal composition holds exactly for unitary monomials, the phase-dressed permutations whose Born kernels realize reversible deterministic state machines. Prescribed sequences evade this obstruction. We classify all pairs of qubit unitary steps, give a necessary-and-sufficient criterion for pairs of qutrit unitary steps, and prove that a boundary-stable unitary sequence on a $d$-dimensional Hilbert space contains at most $d$ fully mixing steps, with equality in every prime dimension. We also construct arbitrarily long genuinely mixing sequences that compose from their initial boundary but fail after an internal restart, and a qutrit-controlled two-qubit realization with active interference. We define a Born–Chapman–Kolmogorov current that vanishes exactly when coherent and stepwise-checked endpoint laws agree, together with an associated measure of how many bits the endpoint reveals about the intermediate checking schedule. This state-machine connection provides a foundation for a quantum theory of music, in which unitary operations are notes and temporal boundaries are cues. Musical-transition prediction is proved PromiseBQP-complete, and the stepwise patterns that preserve a genealogy specify rules for rhythm, whereas the exceptional failure of that genealogy from an internal cue is the quantum music paradox.
Read the full abstract
General Relativity and Quantum Cosmology
Improving the Sensitivity of Gravitational Wave Detection with Weighted Conformal Prediction
In the last decade, kilometre-scale interferometric gravitational-wave detectors have observed hundreds of compact binary mergers, the majority of which are binary black holes. However, the data are noise-dominated, and multiple independent search algorithms (pipelines) are used to enhance sensitivity and improve robustness. Rather than the standard approach of selecting the most significant pipeline output, we combine the outputs from all pipelines using a conformal prediction-based framework to provide statistically rigorous confidence estimates for candidate events. While combining pipelines improves sensitivity and ranking robustness, it requires a principled statistical framework that remains valid as data properties evolve across observing runs. A key challenge is distribution shifts between simulated datasets used for training and calibration and the real, unlabelled, observations used for testing, which can invalidate coverage guarantees and bias confidence estimates. In this work, we address this challenge by incorporating likelihood-ratio reweighting into our conformal prediction framework to account for covariate shift. Using mock datasets containing simulated signals, we demonstrate that weighted conformal prediction restores well-calibrated coverage under covariate shift and increases the confidence of events near the detection threshold, recovering true signals that would otherwise be missed.
Read the full abstract
Systems and Control
Quantifying the Unintentional Islanding Risk: A Comparative Study on Active Distribution Network
The growing share of Inverter-Based Resources (IBRs) is changing the operation of modern Active Distribution Networks (ADNs), raising concerns for grid stability, protection, and reliability. In this paper, we investigate the dynamic behavior of a Medium-Voltage (MV) portion of an ADN during the transition from grid-connected to islanded operation. The study considers a Grid-Following (GFL) converter and Synchronous Generators (SGs) operated under two distinct regulation modes: a fixed-setpoint, non-regulating condition representative of present-day distribution networks, and an active frequency/voltage-regulating condition representative of SGs equipped with governor and AVR control. Detailed electromagnetic transient (EMT) models of both technologies have been developed in DIgSILENT PowerFactory that quantifies the sensitivity of island persistence to SG regulation mode and inertia. The results show that when DERs operate at a fixed power setpoint, the risk of forming a sustained, undetected electrical island is limited, since the isolated network drifts out of the protection thresholds within tens of seconds. Conversely, enabling frequency and voltage regulation on the SGs is sufficient to sustain an unintentional island indefinitely without triggering conventional protection, regardless of system inertia.
Read the full abstract
Computation and Language
TransClean: A Benchmark for Detecting and Extracting Clean Translations from Large Language Model Outputs
Large language models (LLMs) are increasingly used for machine translation, yet their outputs often contain additional text beyond the translation itself, such as language labels, explanations or bilingual repetitions, which we term translation noise. Despite its prevalence, this problem lacks dedicated benchmarks and systematic study. We analyze over 790,000 translation outputs from 12 LLMs across 22 language pairs (LPs) and identify 12 recurring noise patterns, which we group into formatting and content noise. Building on the observed patterns, we construct TransClean, a controlled benchmark of 9,900 pairs of noisy and clean translation outputs, comprising 8,800 synthetically generated instances and 1,100 manually curated authentic instances. We evaluate two extraction approaches on the TransClean benchmark: 1) a span-based extraction method leveraging translation quality estimation models for span detection, and 2) an LLM-based extraction method that prompts an LLM to isolate the translation. Our benchmark and analysis provide the first systematic framework to evaluate and improve the cleanliness of LLM translation outputs.
Read the full abstract
Group Theory
Reversibility and its asymptotic counting in Picard group
We investigate reversible elements in the Picard modular group $\mathrm{PSL}(2,\mathbb{Z}[i])$. We show that reversibility coincides with strong reversibility for Kleinian groups, in particular for the Picard group. We classify reversible elements in the Picard group and characterize loxodromic reversible elements up to conjugacy. We prove that each such conjugacy class contains exactly eight special representatives. We also obtain asymptotic estimates for the number of reversible conjugacy classes with bounded trace.
Read the full abstract
Applied Physics
Influence of magnetic fields on the performance of spin-orbit torque magnetic random-access memory
Spin-orbit torque magnetic random-access memory (SOT-MRAM) offers high speed, ultrahigh endurance, and compatibility with advanced semiconductor processes, making it a promising candidate for next-generation nonvolatile memory. However, intrinsic bias fields in magnetic tunnel junctions (MTJs), originating from reference-layer stray fields and interlayer coupling, cause asymmetric critical switching currents and increased energy consumption. Existing compensation approaches usually introduce additional magnetic layers into the MTJ stack, which increases fabrication complexity and limits wafer-scale integration. Here, we propose a bias-compensation strategy without modifying the MTJ stack by engineering local stray magnetic fields through magnetic filling materials in vertical interconnect access (VIA) channels during the back-end-of-line process. Micromagnetic simulations show that the proposed magnetic filling layer can provide the required auxiliary field for deterministic switching and significantly suppress write-current asymmetry. By optimizing the MTJ position relative to the magnetic filling structure, the write-current bias ratio is reduced from 21.6% in the conventional design to 1.3%. The approach is also applicable to in-plane magnetic anisotropy SOT-MTJs, reducing the bias ratio from 19.8% to -0.2%. Scaling analysis further demonstrates that the compensation effect remains effective when the device size is reduced to 20% of the original dimension (MTJ diameter approximately 10 nm), indicating its potential for high-density SOT-MRAM integration.
Read the full abstract
Algebraic Topology
Homogeneous Milnor fibers and Kato–Matsumoto bounds via simplicial multiwedges
For every $n\geq 3$ and $s\geq 2$, we construct a homogeneous polynomial of degree $n(n+1)/2$ whose Milnor fiber is exactly $2s$-connected and whose rational cohomology contains a strictly defined nontrivial $n$-fold Massey product on classes of degree $2s+1$, implying that the Milnor fiber is non-formal, while attaining the Kato–Matsumoto connectivity bound. Our construction is based on the simplicial multiwedges of the nerve complexes of simple polytopes introduced by Limonchenko, combined with Suciu's realization of weighted homogeneous Milnor fibers. We thereby answer two problems posed by Suciu.
Read the full abstract
Quantum Gases
Microwave-controlled interactions and stripe formation of static-field-shielded polar molecules
We study polar molecules where short-range losses are suppressed by a shielding scheme involving a static electric field and an elliptically polarized microwave field. Using perturbation theory, we derive the effective interaction potential and validate it against coupled channel calculations. We identify a parameter regime where two-body losses are strongly suppressed and the extended mean-field description of dilute molecular Bose-Einstein condensates is justified. We calculate the collective excitations and show that intriguingly, supersolidity in quasi-two-dimensional confinement emerges as a stripe phase even at small values of microwave ellipticity.
Read the full abstract
Statistical Mechanics
Disordered ground states in one-dimensional exactly solvable fluids
One-dimensional fluids of classical hard-core particles of diameter $a$, interacting in pairs via a soft repulsive (monotonically decreasing) potential of finite range $\varphi(x)=\varepsilon \left[ (a'-x)/(a'-a)\right]^{1/\nu}$ $(a\le x\le a')$ with real positive parameters $\varepsilon$ and $\nu$, are studied in an isothermal-isobaric ensemble. If $a'\le 2a$, the pairwise interactions are reduced to nearest-neighbour interactions, which allows for an exact solution of the thermal equilibrium. We focus on the $T\to 0$ ground states, specifically on the equation of state for the mean distance between nearest neighbours $l_0$ and the pair correlation function $g_0(x)$. If $\varphi(x)$ is concave $(\nu\ge 1)$, there exists an “incompressibility” pressure $p_i=\varepsilon/(a'-a)$ such that the ground state is an equidistant chain of particles with spacing $l_0=a'$ for $0<p<p_i$ and with spacing $l_0=a$ for $p>p_i$. If $\nu>1$ (strict concavity), the ground state at $p=p_i$ is disordered with $l_0=\left[ \nu a +(\nu-1)a'\right]/(2\nu-1)$ and $g_0(x)$ being a superposition of weighted Dirac delta functions over discrete positions. If $\nu=1$ (linear ramp), the ground state at $p=p_i$ is disordered with $l_0=(a+a')/2$ and the continuous $g_0(x)$ is a superposition of Heaviside step functions multiplied by polynomials in $x$. The isothermal susceptibility at $T=0$ is nonzero for concave $\varphi(x)$ ($\nu\ge 1$) at $p=p_i$ and, therefore, the corresponding disordered ground states are non-hyperuniform, i.e., they resemble disordered fluids at nonzero temperatures. It turns out that pair correlation functions of disordered ground states, which occur only at a single pressure $p_i$, extend their predictive power to thermodynamic states at nonzero temperatures over a wider range of pressures around $p_i$.
Read the full abstract
Artificial Intelligence
Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning
Test-time adaptation has emerged as a lightweight alternative to costly post-training for improving the reasoning capabilities of Large Language Models (LLMs) on downstream tasks. Predictive entropy provides a model-derived signal for such adaptation, guiding models toward higher-confidence reasoning states without external verifiers or reward models. However, higher confidence does not necessarily imply correctness, as LLMs may remain highly confident along incorrect reasoning trajectories. We observe that high-confidence reasoning is more likely to be correct when confidence remains stable under local perturbations. Based on this observation, we propose Test-Time Adaptation via Stability-Aware Confidence Optimization (TASCO), a framework that incorporates local stability into confidence-based test-time adaptation while keeping the LLM frozen. TASCO operationalizes local stability by optimizing a lightweight task-level prefix under two alternative perturbation strategies: Random Perturbation promotes distributional stability across trajectories induced by nearby perturbed prefixes, whereas Sharpness-Aware Perturbation targets worst-case local sensitivity. Experiments demonstrate that TASCO improves reasoning accuracy and token efficiency across diverse LLMs and reasoning benchmarks, while behavioral analyses show that it maintains stable confidence under local perturbations without prematurely concentrating the model's predictive distribution.
Read the full abstract
Hardware Architecture
PATTON: Enabling Commodity PIM for Production LLM Serving
Processing-in-Memory (PIM) is promising for accelerating memory-bound decode attention, but attention acceleration alone is insufficient for production LLM serving, where engines dynamically allocate, populate, share, cache, and reclaim logical KV cache blocks. Supporting this lifecycle on commodity PIM requires efficient physical memory allocation, block-to-address mapping, and command generation. For the Value cache, these requirements create a fundamental conflict among GEMV efficiency, single-token write efficiency, and memory capacity: GEMV-optimized layouts scatter newly generated Value vectors across rows, making writes costly, while finer-grained memory sharing improves capacity utilization but fragments GEMV reductions. We present PATTON, a PIM runtime that integrates production LLM serving engines with commodity PIM. PATTON introduces hierarchical granule allocation: block-sized Key and Value granules map one-to-one to logical token blocks, fixing their physical placements and commands, while coarser granules group blocks for efficient GEMV execution and memory utilization. A Commit Zone stages partial Value blocks for efficient single-token writes before committing them to GEMV-optimized locations. PATTON tracks these placements to generate KV cache writes and QK-transpose/SV commands. Across attention execution and runtime-induced prefill recomputation, PATTON achieves an average 1.95x speedup and 4.83x higher energy efficiency over evaluated baselines, requires no PIM processing-unit modifications, and maintains a KV cache hit rate comparable to the native GPU KV cache in vLLM.
Read the full abstract
Computers and Society
Buyer Artificial Intelligence-Enabled Environmental Governance and Supplier Environmental Controversies: An Organizational Information Processing and Signaling
Environmental controversies in global supply chains pose significant risks for global buyers. This study examines whether overseas suppliers' exposure to buyers' artificial intelligence (AI)-enabled environmental governance reduces supplier environmental controversies. Drawing on organizational information processing theory and signaling theory, we investigate how suppliers' exposure to AI-enabled governance influences their environmental controversies and the institutional contingencies under which this effect varies. Using text analysis to measure buyer AI-enabled environmental governance, we analyze panel data on 2,505 suppliers of U.S.-listed firms across 41 countries from 2020 to 2024 with multidimensional fixed-effects models. We find that suppliers' exposure to buyer AI-enabled environmental governance is negatively associated with supplier environmental controversies in the following year. This negative relationship is stronger in supplier countries with higher AI readiness and regulatory quality. The study contributes to research on AI-enabled sustainability governance and sustainable supply chain risk management.
Read the full abstract
Information Retrieval
VikingRAG: Accurate and Token-efficient Retrieval-augmented Generation over Structured Documents
State-of-the-art retrieval-augmented generation (RAG) methods exploit document structures to acquire sufficient evidence, but often incur substantial token costs. To reduce structural-context tokens without compromising high RAG accuracy, we present VikingRAG, a directory-aware semantic data management system that tightly integrates semantic and structural access to support structural-context-efficient, evidence-gap-driven multi-round retrieval. To further reduce token overhead of multi-round interaction, we materialize agentic multi-round retrieval traces as experience edges, and reuse these edges for similar queries, avoiding repeated multi-round exploration. To additionally reduce token costs when agentic multi-round retrieval is unnecessary, we introduce an adaptive escalation strategy that answers from one-round experience-augmented retrieval when the evidence is sufficient, and invokes agentic multi-round retrieval only otherwise. Experiments on real datasets show that the base system VikingRAG matches high accuracy of state-of-the-art methods while consuming only 11.6%–51.9% of their tokens. With retrieval-trace reuse and adaptive escalation, token costs drop to 5.1%–32.5% while maintaining competitive accuracy and practical document-storage performance, showing the utility of this work for emerging AI knowledge bases.