Computation and Language
BaatCheet: A Multilingual Corpus for Dialogue Translation in Indian Languages
Priyanka Dasari, Yuvrajsinh D. Bodana, Vandan Mujadia et al. (+3 more)
Existing translation models are typically trained on sentence-level and formal text, limiting their ability to capture everyday conversational dialogue phenomena such as informality, speaker interaction, and discourse coherence. Most existing Indic translation resources and evaluation
benchmarks focus on sentence-level or formal text, making it difficult to assess translation quality of the dialogue phenomena. In this work, we introduce BaatCheet, a
multilingual dialogue corpus named after the Hindi term for conversation or chitchat, containing approximately 49,000 dialogues for dialogue translation across five translation directions. We fine-tune five open-source LLMs across seven training data configurations and find that fine-tuning yields substantial gains over zero- and few-shot baselines. To comprehensively evaluate dialogue translation quality, we employ multiple evaluation strategies, including automatic metrics, LLM-as-judge, and human assessments using an SQM-guided Direct Assessment (DA) Protocol.
Read the full abstract ▾
All authors Priyanka Dasari, Yuvrajsinh D. Bodana, Vandan Mujadia, Arafat Ahsan, Dipti Misra Sharma, Parameswari Krishnamurthy
Artificial Intelligence
TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces
Dehai Min, Daoan Zhang, Yiming Zeng et al. (+13 more)
An
agent can complete a task while exhibiting undesirable behavior during execution. Developers need tests for the specific behaviors encountered in deployment, beyond fixed
benchmark suites. We present TraceDance, an
agent system that constructs targeted
benchmarks from deployment traces for user-specified undesirable behaviors. For efficient construction, Anchor-and-Confirm combines programmable
retrieval with candidate-level confirmation by a Flash large
language model (LLM), while the Anchor Synthesis Loop generates and revises specifications for custom behaviors. The
benchmarks use decision-point continuation to evaluate an LLM's next turn at a recorded decision point with a behavior-specific rubric, without a reference answer or environment replay. Experiments in coding and general tool use draw on 252,557 sessions and produce 107
benchmarks with 4,125 instances, fulfilling 95.3% of build-target requests. Both human annotators confirm the requested behavior in 84% of sampled instances, and the automated grader's agreement with human pass/fail judgments is comparable to that between the annotators. Nine frontier LLMs achieve a mean pass rate of only 26.7%, showing that they still struggle to respond appropriately at the evaluated decision points. Analysis across behavior-specific
benchmarks further reveals weaknesses in how current LLMs behave as
agents. By turning deployment problems into targeted
benchmarks, TraceDance could serve as a key component of the recursive self-improvement (RSI) loop.
Read the full abstract ▾
All authors Dehai Min, Daoan Zhang, Yiming Zeng, Huayi Zhang, Ziyi Chen, Yan Zhang, Qinbo Bai, Mengyuan Chao, Jing Ning, Qiyue Hua, Huiyi Chen, Hanrong Zhang, Henry Peng Zou, Jie Yang, Wei Xu, Philip S. Yu
Optics
Tailored multipolar optical trapping with higher-order structured light
Nabendu Mishra, Nick Vamivakas and Eileen Otte
Dipolar optical forces can confine Rayleigh-sized dielectric particles towards or away from intensity maxima, commonly known as bright and dark trapping. Structured light illumination opens up opportunities for customized trapping landscapes, unveiling the possibility of gray trapping, especially for larger Mie particles that can excite a higher-order multipolar response. While previous studies have primarily focused on calculations of the gradient force for some exemplary, focused scalar beams, we not only extend these analyses by combining vectorial focusing with generalized Lorenz-Mie theory, but also consider more complex trapping fields. Motivated by their unique tight focusing behavior, shaping strong transverse intensity gradients and three-dimensional electric field components, we explore higher-order cylindrical scalar and vector beams. We theoretically investigate multipolar trapping of Mie-regime silicon nanospheres in water, calculating the three-dimensional net radiation force and assess local trapping stability through the force Jacobian. Near an octupolar Mie resonance, we predict stable gray trapping with enhanced efficiency for vector beams. Further, we find field geometries enabling configurable, particle-size-dependent gray/dark trapping with subwavelength inter-particle spacing. These results establish polarization, illumination structure and particle parameters as complementary controls for radius-dependent multipolar trapping, providing a basis for potential applications in sub-wavelength particle sorting, trapping-assisted nanopatterning and optical binding investigations.
Read the full abstract ▾
Statistics Applications
Timing of First Alzheimer's Neuroimaging: A Multicenter Analysis of Genetic, Cognitive, and Social Factors
J. T. Korley and E. Adu Bonsu
Background.
Magnetic resonance imaging (MRI) is central to Alzheimer's disease research and clinical evaluation, yet timing of first MRI after cohort entry varies substantially. In aging cohorts, many participants die before imaging, and treating death as noninformative censoring can distort inference on neuroimaging access and conceal disparities. Objective. To quantify individual-level determinants and center-level heterogeneity in timing of first neuroimaging while accounting for the competing risk of death. Methods. We analyzed National Alzheimer's Coordinating Center Uniform Data Set v3 linked to MRI records from 2015 through March 2025. The study included 20,867 participants aged 50-95 years without dementia at baseline and with known APOE {\epsilon}4 status. Time from baseline to first post-baseline MRI was the primary event, with death before MRI treated as a competing event. We fit
Bayesian multilevel cause-specific accelerated failure time models with Alzheimer's Disease Center random intercepts, using Weibull and log-logistic models for MRI timing and death, respectively. Frequentist accelerated failure time and Cox models were used for sensitivity analyses. Results. APOE {\epsilon}4 was not credibly associated with timing of first MRI. Older age, lower educational attainment, and non-White race were associated with delayed MRI acquisition. Center-level variation in MRI timing was substantial. Conversely, APOE {\epsilon}4 homozygosity was associated with earlier death prior to MRI, alongside strong effects of age and cognitive impairment. Conclusions. Timing of first MRI in Alzheimer's research cohorts is driven primarily by demographic, social, and center-level factors rather than genetic risk. Accounting for competing mortality and site heterogeneity is necessary to avoid biased neuroimaging samples and support equitable Alzheimer's research.
Read the full abstract ▾
Data Structures and Algorithms
On the Complexity of Forcing and Anti-Forcing Minimum Cuts
Tatsuya Gima, Yasuaki Kobayashi, Hiraku Morimoto et al. (+1 more)
For an instance of a combinatorial optimization problem, a forcing set is a set of elements such that there is a unique optimal solution including it. Symmetrically, an anti-forcing set is a set of elements such that there is a unique optimal solution excluding it. In this paper, we study the problems of computing smallest forcing and anti-forcing sets for two classical cut problems, Global Min Cut and \textsc{Min $s$–$t$ Cut}. We also consider variants in which the optimal cut to be uniquely determined is given as input. For each of these problems, we either give a
polynomial-time
algorithm or prove \NP-completeness.
Read the full abstract ▾
All authors Tatsuya Gima, Yasuaki Kobayashi, Hiraku Morimoto, Yota Otachi
Machine Learning (Statistics)
The Statistical Benefits of Multiple Responses for Learning from Demonstrations
Chandramauli Chakraborty and Cong Ma
Many generative systems return multiple candidate responses and are evaluated according to the best one. Recent work shows that, when demonstrations are optimal, pass@$k$ can reduce the sample complexity of learning from demonstrations by a logarithmic factor in $k$. We ask what happens when the demonstrator is not assumed to be optimal. We find that multiple responses provide a qualitatively stronger benefit in this setting. In a finite reward-class model with no reward feedback, moving from pass@$1$ to any pass@$k$ with $k\ge2$ changes the worst-case dependence on target accuracy from $1/\varepsilon^2$ to $1/\varepsilon$, uniformly over demonstrator quality. Under standard evaluation, where an unknown reward is fixed before training, increasing $k$ provides an additional and distinct benefit: the optimal dependence on a reward class of size $N$ improves from $\log N$ to $\log N/\log k$. We further show that these two effects can be separated. Under robust evaluation, where one learned policy must compete with the demonstrator simultaneously for every reward in the class, the fast $1/\varepsilon$ dependence persists, while the $1/\log k$ improvement can disappear. We establish matching upper and lower bounds in the corresponding regimes and give a greedy multiplicative-weights learner achieving the upper bounds without any assumption on demonstrator quality.
Read the full abstract ▾
Computation and Language
Calibration, Not Answer Selection: Distilling Internal Confidence in Reasoning Models
Yadong Xi, Rongsheng Zhang, Tangjie Lv et al. (+2 more)
Reinforcement learning with binary correctness rewards trains correctness, not calibrated confidence. The confidence that
reasoning models verbalize is systematically overconfident, and the problem is not merely one of scale: verbalized confidence tracks how willing a model is to commit to an answer, not how likely the answer is to be right. Post-hoc rescaling therefore fits one distribution but rarely transfers. We look inside the model instead. On factual question answering, a linear probe on the hidden state between the chain of thought and the answer is substantially better calibrated: its expected calibration error is 5 to 38 times lower than that of the verbalized score across four
benchmarks and two model families. However, when used to pick among N sampled answers, that same probe nearly ties majority voting yet falls far short of the oracle. Internal states answer "how certain am I" well and "which answer is right" poorly, so the signal should be reported as a confidence rather than used to select answers. As a result, we introduce probe-guided self-
distillation (Probe-SD): score a model's own sampled traces with the probe, overwrite the confidence each trace states, and finetune the base checkpoint of the same family, so nothing but the model itself remains at test time. On Qwen3-14B, Probe-SD cuts ECE from 0.178 to 0.024 in-domain and from 0.542 to 0.113 out-of-domain, where it also beats post-hoc recalibration and self-consistency
distillation. The resulting confidence is well-calibrated and useful for weighted voting, behaviors previously attributed to online RL, here obtained with supervised finetuning alone.
Read the full abstract ▾
All authors Yadong Xi, Rongsheng Zhang, Tangjie Lv, Ziyang Luo, Ruochen Zhao
Artificial Intelligence
Learning to Sell: Reinforcement Learning for Strategic Large Language Model Agents in Multi-Product Markets
Shuze Daniel Liu, Claire Chen, Jiuqi Wang et al. (+2 more)
Autonomous large
language model (LLM)
agents operating in multi-product
markets must make sequential decisions under information asymmetry and resource constraints. We develop a machine learning approach for training such
agents to act effectively as sellers in a multi-item bargaining environment, where a seller concurrently negotiates a catalog of substitutable assets across a pool of independent buyers. Buyers hold private, heterogeneous valuations across products, and each can purchase at most one item. Facing limits on total communication turns, the seller must dynamically match buyers with the most profitable products considering their private valuations, while strategically allocating its limited interaction budget toward combinations of greater potential value. We formalize this problem as a Partially Observable Markov Decision Process using a structured, four-part message protocol that maps natural
language into a parsable and regulated decision space. Using this formalization, we design a post-training method using Reinforcement Learning from Verifiable Rewards (RLVR). To evaluate this framework, we construct a multidimensional metric suite that quantifies constraint adherence, seller surplus extraction, and allocation quality. Our trained seller
agent learns to match limited inventory to buyers more effectively, matching or outperforming trillion-parameter frontier models in both seller surplus extraction and buyer-product allocation quality. Finally, these learned strategies generalize robustly to unseen
market structures, correlated valuation distributions, and price ranges not encountered during training.
Read the full abstract ▾
All authors Shuze Daniel Liu, Claire Chen, Jiuqi Wang, David Simchi-Levi, Thorsten Joachims
Computer Vision
Informative Viewpoint Selection for Episodic-Memory Embodied Question Answering using Omnidirectional Images
Kaname Kitamura and Asako Kanezaki
Embodied Question Answering (EQA) requires
agents to answer natural
language questions about surrounding environments from visual observations. In this work, we focus on open-vocabulary episodic-
memory EQA (EM-EQA), where an
agent answers free-form questions using recorded observation histories. Omnidirectional images are promising for this task, as they provide wide field-of-view observations that can capture surrounding context without requiring explicit camera rotations. However, omnidirectional images introduce two challenges for EQA: (i) equirectangular projection causes severe
geometric distortion that degrades
vision-
language model (VLM) recognition accuracy, and (ii) feeding equirectangular images directly into VLMs introduces excessive irrelevant background information, reducing answer accuracy and increasing the visual-
token burden. To address these challenges, we propose a viewpoint selection method for EM-EQA using omnidirectional images. Our method converts equirectangular observations into perspective views via cubemap projection, estimates question-conditioned relevance with fine-tuned BLIP-2, and selects informative and diverse viewpoints through diversity-aware greedy selection. Experiments on the Habitat-Matterport 3D (HM3D) subset of OpenEQA show that our method achieves state-of-the-art model performance among the reported model results with equirectangular observations. Moreover, after removing rotation views, which reduces observation frames by 65.5%, our method largely maintains its answer accuracy.
Read the full abstract ▾
Artificial Intelligence
Feedback Makes Perfect: A Closed-Loop Framework for NL-to-STL Translation
Bowen Ye and Xiang Yin
Signal Temporal Logic (STL) enables rigorous verification and control of cyber-physical systems, but writing correct specifications requires expertise that most requirement holders lack. Large
language models can translate natural-
language (NL) requirements into STL, yet stronger translators alone approach an accuracy ceiling. We argue that this ceiling stems from how the task is posed: one-shot, open-loop translation is somewhat ill-defined. Natural
language is ambiguous, and, more fundamentally, what a person writes may not always be what they intend, so the target specification is not fully contained in the input text. We therefore reformulate NL-to-STL translation as a closed-loop feedback process. Each generated formula is translated back into natural
language for the user to check, and natural-
language corrections drive revision until the user accepts the specification. Users never read or write formal syntax. This framework rests on an asymmetry familiar from feedback control theory. The forward path, from ambiguous
language to formal logic, is hard and error-prone. The feedback path, from structured STL back to
language, can be made highly precise, and a precise feedback path lets an imprecise forward path achieve precise closed-loop behavior. Experiments on 500 expert-authored requirements and seven LLMs support this view. Back-translated explanations agree with expert judgments in 99.5% of cases. Closed-loop refinement raises strong models from about 89% open-loop accuracy to 98.0–99.2%, and yields gains of over 30 percentage points for weaker models (e.g., 17.6%$\rightarrow$48.0%). Ablations show these gains come from the semantic content of the feedback rather than from repeated attempts. An expert audit and a 280-session user study further confirm the reliability of the loop. We also identify a capability threshold above which feedback no longer helps.
Read the full abstract ▾
Computer Vision
InfoEdit: Probing Global Layout Reasoning in Infographic Editing
Cheng Yang, Chufan Shi, Huijuan Wang et al. (+6 more)
Multimodal foundation models edit natural photographs at production quality, yet the same models struggle with structured visual content such as infographics. Unlike photographs, infographics encode information through logical relations; editing one element often requires surrounding elements to be adapted. We refer to this global layout
reasoning capability as reflow. Existing image-editing
benchmarks neither provide a dedicated setting for structured visual content nor evaluate the reflow capability. We introduce InfoEdit, a novel
benchmark of 1,000 infographics across eight logical-relation families, paired with 4,000 editing instructions across four editing tasks, and a reflow-aware evaluation protocol. Across eight frontier editors, only GPT-Image-2 clears 60% average success rate; most models fall below 7%, and no editor exceeds 36% on the Swap-Block task even with perfect target localization. We further show that code-level editing can match the strongest pixel-level editor, revealing complementary strengths across tasks. InfoEdit identifies reflow as a central challenge in structured visual content editing and provides a diagnostic
benchmark to facilitate future progress.
Read the full abstract ▾
All authors Cheng Yang, Chufan Shi, Huijuan Wang, Bo Shui, Yaokang Wu, Muzi Tao, Yibo Yan, Xuezhe Ma, Taylor Berg-Kirkpatrick
Plasma Physics
In-situ Generation of Polarized Attosecond Positron Bunches via Radiation-Wakefield Induced Two-Photon Pairs
Zhen-Ke Dou, Qian Zhao, Hao-Tian Wang et al. (+2 more)
High-energy, dense,
spin-polarized positron beams with attosecond duration are highly desirable for advanced accelerator physics, laboratory astrophysics, and ultrafast matter-antimatter studies, yet remain beyond the capability of conventional source and injection-based
plasma-acceleration schemes. We propose an integrated in-situ method for production, trapping, and accelerating polarized attosecond positron bunches in a single
plasma stage. An ultraintense hollow
laser propagating in a
plasma channel drives a radiative wakefield, where longitudinally injected electrons emit a dense MeV gamma-ray bath through nonlinear Compton scattering. Subsequent
photon-
photon collisions induce linear Breit-Wheeler pair production, while linear and nonlinear radiative processes mediate polarization transfer to the produced positrons.
Spin-resolved QED particle-in-
cell simulations indicate that the
laser-modulated wakefield naturally captures positrons born in the frontier bubble wake, forming collimated bunch trains with hundred-attosecond duration, GeV-level energies, and densities up to 10^17 cm^-3. Parameter scans demonstrate robustness against variations of
laser intensity and
plasma-channel density. This single-shot, source-free approach simultaneously addresses positron production, polarization, temporal compression, and
plasma injection, offering a compact route toward polarized attosecond antimatter beams.
Read the full abstract ▾
All authors Zhen-Ke Dou, Qian Zhao, Hao-Tian Wang, Feng Wan, Jian-Xing Li
Artificial Intelligence
RINI: Seeing the Prior Is Not Enough
Hongyi Du, Tianyi Zhang, Heng Wang et al. (+7 more)
A research proposal can describe an established mechanism correctly while claiming to introduce it. We study whether providing the earlier paper corrects such contribution claims. Three controlled experiments compare proposals generated with a contribution-bearing prior and a same-topic control. Providing the prior yields no clear aggregate reduction in unsupported novelty. Human analysis of 175 interpretable exposed proposals finds that 137 recognize the prior's relevance, but 61 correctly attribute the established contribution. Of 71 proposed remaining distinctions, 37 are covered by the same prior. We introduce Research Idea Novelty Inspection (RINI), which audits contribution claims against evidence, checks the remaining distinction, and applies local revisions. Five human annotators evaluate 1,080 original-revision pairs across three methods. On the same 240 originals judged to require correction, successful repair is 11.7% for Self-Revision, 39.1% for Retrieve-and-Revise, and 72.2% for RINI, with research tasks weighted equally. The improvement over same-evidence direct revision is 33.0 percentage points. The revised proposals retain their research questions and technical methods. These results motivate explicit contribution attribution when using literature to generate and revise research proposals.
Read the full abstract ▾
All authors Hongyi Du, Tianyi Zhang, Heng Wang, Zhelun Gao, Yimei Liu, Ambrose Luo, Annie Hao, Jiayan Ni, Jiawei Han, Jiaxuan You
Classical Analysis and ODEs
Weighted Bernstein-type inequalities on a trapezoidal domain
Tomasz Beberok
We study weighted $L^2$ Bernstein-type inequalities for algebraic
polynomials on a trapezoidal domain. Using a parametrization by the unit square, we construct a natural symmetric second-order Sturm–Liouville operator and derive exact directional Bernstein inequalities with explicit constants. We also obtain strengthened estimates for the ordinary partial derivatives. Finally, we relate the corresponding boundary factors to Dini derivatives of the Siciak extremal function of the trapezoid.
Read the full abstract ▾
Artificial Intelligence
Multi-Dimensional Comparative Scale Construction for Efficient Personalized Subjective Judgment in High-Traffic Applications
Xianglong Shi, Shifeng Liu, Sirui Zhao et al. (+2 more)
Subjective judgments are central to many high-traffic applications, but subjective intensity is difficult to quantify and perceptions vary substantially across individuals. To address these challenges, we propose a pairwise comparative framework for multi-dimensional scale construction. By comparing case-person pairs along case and profile dimensions, the framework constructs relative scales that capture both fine-grained intensity and individual variation. To support practical high-traffic deployment, we optimize both offline scale construction and online inference. For scale construction, we combine sparse Elo comparisons with multi-judge voting, cutting the comparison cost from $O(N^2)$ to $O(NK)$ for $N$ objects and a budget of $K$ opponents per object, while limiting reliance on any single judge. For inference, we propose SubJudge, a System One model for personalized scoring with Batchwise Preference Optimization (BPO). Using Bradley-Terry comparisons, BPO trains the model to learn relative orderings, and SubJudge reads a continuous score from digit-
token probabilities at the first response position, requiring only one forward pass per criterion and reducing the inference complexity to $O(1)$. Experiments on PluriHarms and iNews show that our 9B models match or surpass the evaluated frontier LLMs on multiple metrics. On the H100 GPU, SubJudge achieves an approximately $1.29\times$ to $261\times$ speedup in mean inference latency over Qwen3.5-9B with different thinking budgets. The code is available at https://github.com/Longchentong/SubJudge.
Read the full abstract ▾
All authors Xianglong Shi, Shifeng Liu, Sirui Zhao, Shengming Yuan, Enhong Chen
Cryptography and Security
Application Agnostic EM Side-Channel Emanations of the FPGA Clock Distribution Network
Ashish Sharma, Alia Long, Keith D. Mecham et al. (+1 more)
Electromagnetic side channels have been extensively utilized in non-invasive attacks or analyses to extract critical information from deployed systems. Most of the attacks or analyses are typically executed on cryptographic
algorithms under the assumption that the underlying implementation remains fixed. The assumption is valid in the context of fixed hardware, where modifications to the
circuit are not made after manufacturing. However, in the case of adaptive hardware, including field-programmable gate arrays (FPGAs), the assumption no longer holds as modifications to the configuration of the array are permitted post-manufacturing. Consequently, this study introduces a methodology for the analysis of electromagnetic side channels of an FPGA through the study of deterministic finite state machines (DFSMs), where the effects of placement, routing, and operating frequency on the EM emanations of the FPGA due to the utilized clocking resources are characterized. To analyze the impact of frequency, device placement, and resource allocation on the EM emanations from the FPGA, statistical analysis including mean difference and variance difference calculations of measured EM fields were utilized to identify points of high EM activity on a set grid of sampled points. A 1.32x increase in flagged grid points was observed in the analysis of the mean difference when the percentage of utilized clock resources was increased from 25% to 90%. Similarly, a 1.39x increase in flagged grid points was observed when the variance difference was calculated for the same increase in clock resources.
Read the full abstract ▾
All authors Ashish Sharma, Alia Long, Keith D. Mecham, Ioannis Savidis
Computation and Language
The Text Beside the Image: Detection, Utility and Leakage for Trustworthy Multimodal Medical Data and Beyond
Andreas Maier, Monica Hinrichs-Mayer, Franziska Weber et al. (+4 more)
Medical images are released with the reports that describe them, and protecting the image does not protect the report. This paper measures the text component of such releases. We measure identifier detection, downstream utility and residual identity leakage on the same documents, with the pseudonymisation policy as the variable under test: 15 detectors, three release conditions and four corpora of medical reports, legal judgments, news and other genres, and e-mail, in German, English, Chinese and Arabic. A fixed 13-detector union reaches a person sensitivity of 0.9998 at specificity 0.8686 on the medical reports, 0.9958 at 0.8504 on the legal judgments, 0.9352 at 0.9318 on news and other genres, and 0.9906 at 0.6235 on e-mail. With this ensemble, frequency matching with a public name list recovers zero identities by
alignment across the four corpora; the names it got right were ones the detector missed, left in clear text. Cross-document linkage ranks the correct person first for 0.93% of e-mail queries without training and 3.94% with it, against 1/3697 chance and 71.98% on unmodified text. On the medical reports it recovers nothing without training and 0.71% of 138 queries with it, against 1/207 chance and a 2.73% ceiling on unmodified text.
Read the full abstract ▾
All authors Andreas Maier, Monica Hinrichs-Mayer, Franziska Weber, Niklas Lackner, Matthias May, Bernhard Kainz, Siming Bayer
Machine Learning
Domain Generalization under Sampling Pattern Shifts in Irregular Time Series
Changhun Kim, Joohyung Lee, Kwanhyung Lee et al. (+3 more)
Irregularly sampled multivariate time series (ISMTS) are prevalent in real-world applications, where both observation times and available measurements can vary substantially across domains. While recent models increasingly exploit such sampling information for prediction, its robustness under sampling pattern shifts remains underexplored. We introduce HAR-C, to the best of our knowledge the first controlled
benchmark for sampling pattern shifts in ISMTS, and show that sampling shifts alone can substantially degrade performance, induce sampling-specific shortcuts, and remain challenging for existing domain generalization (DG) methods. Motivated by these findings, we propose PRISM, a DG framework that first learns complementary feature-centric and sampling-centric representations without task labels, and subsequently performs robust supervised training across diverse sampling variations to discourage brittle shortcut reliance. Extensive experiments on controlled and real-world ISMTS
benchmarks demonstrate that PRISM consistently improves robustness to unseen sampling shifts over existing methods. Our code is available at https://anonymous.4open.science/r/PRISM.
Read the full abstract ▾
All authors Changhun Kim, Joohyung Lee, Kwanhyung Lee, Donghwee Yoon, Grigorios Chrysos, Eunho Yang
Number Theory
Prime-Detecting Identities from Dirichlet Inversion
Zhichen Liu
For each integer k >= 1, let sigma_k(n) = sum_{d|n} d^k, let sigma_k^{-1} denote its Dirichlet inverse, and let J_k denote the kth Jordan totient function. Using established
prime-power values of sigma_k^{-1}, we first obtain the elementary criterion sigma_k^{-1}(n) + J_k(n) + 2 = 0 precisely when n is
prime. At k = 1, this involves Euler's totient J_1 = phi and provides a
prime-only comparison for the next equation. Our main result classifies the zeros of sigma_k^{-1}(n) + mu(n) + J_k(n) + 3: for k = 1, they are precisely the
primes and 18; for k >= 2, they are precisely the
primes. The proof separates integers according to their
prime-exponent patterns. Sign and divisibility arguments exclude composite zeros outside the cube-free nonsquarefree case, where an additional divisibility condition isolates 18 at k = 1 and precludes composite zeros for higher k. Standard convolution and Lambert-series identities provide the arithmetic framework, while an intermediate-divisor relation explains the exceptional zero. These criteria are structural characterizations rather than efficient primality tests, since direct evaluation of the multiplicative formulas generally presupposes a factorization of n.
Read the full abstract ▾
Quantum Physics
Preparation-controlled relaxation in an overdamped RLC circuit: A pedagogical route to the spectral Mpemba effect
Matheus H. dos Santos, Cler T. Garcez, Jeveson C. da Silva et al. (+2 more)
Can a system with more stored energy cool faster than one with less? An overdamped resistor–inductor–capacitor (RLC)
circuit offers a classical, experimentally accessible spectral Mpemba analogue: both states lose energy monotonically, yet the initially higher-energy state can cross below the lower-energy one. The reason is that initial energy alone does not determine the relaxation path; initial charge and current set how strongly the slow and fast decay modes are excited. We develop this mechanism from deterministic to
stochastic and open-system descriptions. For the classical
circuit, we derive the modal preparation conditions and crossing criterion, explain geometrically why equal-energy states can relax differently, and confirm the effect experimentally by reconstructing charge, current, and energy from oscilloscope traces. The inversion survives when the
circuit is driven by externally injected white noise. An
equilibrium Ornstein–Uhlenbeck treatment separates information carried by the ensemble mean from that stored in its covariance: centered Gibbs preparations preserve mean-energy ordering, while displaced or anisotropic Gaussian states can encode distinct slow and fast sectors. A Fokker–Planck generator then bridges to Markovian
quantum dynamics. For a thermally damped
quantum harmonic oscillator, the mean bare-energy excess decays with a single exponential factor, forbidding energy-order inversion for finite mean occupation. A below-threshold parametric oscillator restores slow and fast quadrature sectors, allowing crossings controlled by displacement and covariance. The Liouvillian formulation shows that observed relaxation depends jointly on the generator spectrum, the modes populated by preparation, and the modes visible to the observable.
Read the full abstract ▾
All authors Matheus H. dos Santos, Cler T. Garcez, Jeveson C. da Silva, G. D. de Moraes Neto, Norton G. de Almeida
Artificial Intelligence
ChronoFlow: Hierarchical Flow Matching for Irregular Time Series Generation
Changhun Kim, Sunguk Jang, Jeongjun Lee et al. (+5 more)
Recent advances in generative modeling have substantially improved time series generation, yet most existing methods either assume a regular temporal grid or focus on feature dynamics under a given sampling structure. This makes them illsuited for generating irregular time series in their native form, where a model must capture not only feature values, but also how many observations occur, when they occur, and which features are observed together. To address this heterogeneous generation problem, we propose ChronoFlow, a unified hierarchical flow matching framework organized by statistical granularity. Following a coarse-to-fine hierarchy, ChronoFlow first generates observation counts and feature-wise frequencies, then jointly generates observation times and feature co-observation patterns, and finally generates values conditioned on the realized pattern. This turns a complex joint generation problem into structurally aligned subproblems while preserving their dependencies. To evaluate complete irregular time series generation, we introduce complementary metrics spanning sample realism, sampling structure, value fidelity, and temporal and cross-feature dependencies, and validate them through controlled corruptions. Across five
benchmarks, ChronoFlow achieves strong improvements in generation fidelity over existing baselines, while factorization studies support the proposed hierarchy. Our code is available at https://anonymous.4open.science/r/ChronoFlow.
Read the full abstract ▾
All authors Changhun Kim, Sunguk Jang, Jeongjun Lee, Juhwan Choi, Sangchul Hahn, Grigorios Chrysos, Eunho Yang, Juho Lee
Quantum Gases
Mobile-impurity dynamics in a non-Hermitian Stark-localised bath
Muhammad S. Hasan, Hsiang-Hua Jen and Jhih-Shih You
We investigate the dynamics of a clean mobile
impurity coupled to a bath subjected to a linear Stark potential as well as asymmetric hopping in one dimension. The
impurity mediates resonant tunnelling in the bath whenever the interaction strength matches the Stark potential and leads to loss in the
memory of the initial state of the bath particles. The non-Hermitian asymmetry does not shift these resonances but reshapes them directionally, and it destroys the bath's Stark-induced
memory through the non-Hermitian skin effect. On the other hand, we find that the
impurity is localised only within an intermediate range of Stark potential. At very weak tilt strengths the bath offers no potential to trap the
impurity, while at strong tilt the bath freezes into a nearly periodic charge density wave that pushes the
impurity towards partial localisation. Counterintuitively, the
impurity localises in the intermediate regime where the bath's initial-state
memory is strongly suppressed, yet a spatially inhomogeneous density profile persists. The interaction amplifies these density fluctuations, dressing the
impurity with an effective disordered potential that traps it. Since the skin effect degrades this intermediate density distribution, the
impurity localisation window shifts toward stronger tilt in the non-Hermitian regime.
Read the full abstract ▾
Computational Engineering
Modelling non-linear aeroelastic loads in long-span bridges with extreme learning machines
Gledson Rodrigo Tondo, Samir Chawdhury, Sergio Andres Castro Giraldo et al. (+1 more)
Accurate modelling of aerodynamic loads is essential for predicting instabilities and ensuring the safety of long-span bridges. A methodology is introduced for modelling aerodynamic self-excited forces in bridge-deck cross-sections using extreme learning machines (ELMs). ELMs, as single-layer feedforward neural networks, offer efficient training and accurate predictions. Forced-oscillation
datasets from computational fluid dynamics (CFD) or wind-tunnel experiments are used for training, enabling systematic data selection to capture non-linear aerodynamic behaviour often missed by semi-analytical approaches. Once trained, the model predicts self-excited loads for any arbitrary motion composed by frequencies and amplitudes within the training domain. Comparisons with analytical, semi-analytical, and CFD results show superior accuracy in capturing non-linear force components and close agreement for aerodynamic loads and flutter wind speeds. Training required about 1.1% of the time of a conventional neural network, and coupled flutter analysis runs in seconds, providing orders-of-magnitude speed-ups over CFD. These results indicate that ELM-based frameworks are accurate, practical, and efficient alternatives for modelling self-excited loads, particularly when preliminary CFD or wind-tunnel data are available. The presented approach offers a reliable data-driven technique for aeroelastic load modelling in long-span bridges.
Read the full abstract ▾
All authors Gledson Rodrigo Tondo, Samir Chawdhury, Sergio Andres Castro Giraldo, Guido Morgenthal
Machine Learning
Towards Identifiable Representations under Misspecified Structure
Yuke Li, Yujia Zheng, Ziyi Chen et al. (+2 more)
The presence of noise that depends on the latent variables poses a fundamental challenge to identifiability. Existing results rely on conditional independence among the observations given the latent variables. We study a more general misspecified structure, where this conditional factorization does not hold, and establish both precise and approximate identifiability guarantees. We characterize structural misspecification as a perturbed factor analysis problem. For precise identifiability, we establish subspace identifiability under spectral separation and controlled perturbation, followed by component-wise identifiability under structural sparsity. When the precise condition is not guaranteed, we derive an approximate subspace-identifiability
theorem. Based on these results, we develop an unsupervised variational
estimator for recovering latent variables. Experiments demonstrate the effectiveness of the proposed framework.
Read the full abstract ▾
All authors Yuke Li, Yujia Zheng, Ziyi Chen, Kun Zhang, Heng Huang
Quantum Algebra
Classifying Grothendieck rings of pivotal fusion categories of rank four
Jingcheng Dong, Sebastien Palcoux and Arnaud Plessis
We classify the Classifying Grothendieck rings of pivotal fusion categories of rank four over the complex field. There are exactly fifteen, each admitting a
unitary categorification. Central induction and Frobenius-Schur indicators give a uniform Frobenius-Perron dimension bound of 3600. The finite classification combines new arithmetic and twist obstructions with an exhaustive census and exact exclusion certificates.
Read the full abstract ▾
Artificial Intelligence
Next Thoughts Are Distributions: Generative Autoregressive Reasoning in the Latent Space
Yang Li, Yi Wang, Shiyuan Huang et al. (+3 more)
Reasoning problems often admit multiple valid ways to proceed. Continuous
reasoning promises to move computation beyond
language tokens into a more compact latent space, but representing several plausible ways to think next remains difficult. We introduce Autoregressive Thought Flow (ATF), which models the next continuous thought as a
multimodal distribution. A causal autoregressive model performs the
reasoning computation, while a lightweight
diffusion head generates a plausible next thought from the resulting condition. The sampled thought is fed back into the model, allowing continuous
reasoning to unfold for a variable number of steps while preserving the pretrained backbone. Across mathematical
reasoning tasks, ATF improves accuracy with compact latent traces and benefits from reinforcement learning and additional test-time thinking. Multi-sample evaluation shows broader solution coverage, indicating that its
multimodal predictions capture useful diversity among
reasoning paths. Our results suggest that continuous
reasoning is more effective when multiple possible next thoughts remain available rather than being collapsed into a single prediction.
Read the full abstract ▾
All authors Yang Li, Yi Wang, Shiyuan Huang, Yang Liu, Hao Wang, Chengzhi Mao
Artificial Intelligence
Structured Sparse Memory for Recurrent Reasoning
Zixuan Zhao, Samuel Wheeler, Neil Getty et al. (+3 more)
Recurrent models trained from scratch have recently become competitive on ARC-style
reasoning tasks, but the usual framing around small recurrent backbones overlooks two important parts of the system: task-conditioned
memory and synthetic augmentation data. We study this regime through CHARM, a compact hybrid ARC model that combines recurrent
reasoning with structured task
memory, synthetic data, and inference-time aggregation. In existing approaches, task-conditioned
memory supplies a large hidden source of capacity, reaching more than 30x the size of the recurrent backbone. We introduce a compositional sparse
embedding (CoSE) for task conditioning that reduces learned task-
memory parameters by over 90% while improving pass@2 in controlled ARC ablations. We also introduce Re-ARC2, a set of programs that generate and verify new examples for all ARC-AGI-2 training tasks. Combining these ingredients, our system reaches 84% pass@2 on ARC-AGI-1 and 46.7% pass@2 on ARC-AGI-2 public evaluation. Structured
memory also enables adaptation with only 512 parameters per unseen puzzle, and its benefits generalize to other domains. Our code,
dataset, and model checkpoints are available at https://github.com/water-vapor/charm.
Read the full abstract ▾
All authors Zixuan Zhao, Samuel Wheeler, Neil Getty, Xiaotian Duan, Rick Stevens, Fangfang Xia
Robotics
Q-WAM: 4-Bit Quantization of World Action Models with Action-Subspace Protection
Arash Akbari, Arman Akbari, Jingwu Luo et al. (+7 more)
World Action Models (WAMs) jointly generate
video and
robot actions through iterative
diffusion and perform strongly in robotic manipulation. However, their prohibitive compute and
memory costs pose substantial deployment challenges. Post-training quantization (PTQ) can reduce these costs, but existing PTQ methods such as smoothing and rotation are insufficient to maintain the precision of action generation. To overcome this limitation, we propose Q-WAM, a new 4-bit weight-activation quantization for WAMs that preserves the actions the model generates. Specifically, we introduce the Action Observability Gramian (AOG), which measures how much rounding errors in each weighted combination of a layer's input channels change the final action through all denoising steps. We also develop Action-Subspace Protection (ASP), which keeps the few most action-sensitive channel combinations in a tiny 16-bit low-rank branch and quantizes the complementary weights and activations to 4 bits, both as dense
matrix multiplications that run efficiently on GPUs. Finally, to preserve action quality with minimal overhead, we identify the experts that matter most for the generated action by aggregating the AOG-derived action mass across the layers of each expert and apply ASP only to those experts. We evaluate Q-WAM on three WAMs, both in simulation and in real-world deployment. On the RoboTwin 2.0
benchmark, it reaches 89.6–93.0% average success rate, within 1.1 percentage points of the 16-bit models, while reducing the
memory of the quantized blocks by 3.1–3.4$\times$. Our method outperforms the strongest baseline, SVDQuant, by 2.5–8.7 percentage points. On a Unitree G1
humanoid and a bimanual UR3
robot, it improves success over SVDQuant by 12.8-17.6 percentage points.
Read the full abstract ▾
All authors Arash Akbari, Arman Akbari, Jingwu Luo, Yuhao Lei, Yi Gao, Weiwei Chen, Xuan Zhang, Zhenman Fang, Geng Yuan, Yanzhi Wang
Artificial Intelligence
LSTMem: Hierarchical Long Short-Term Online Memory for Large Language Models
Xianglong Shi, Ruijie Yang, Sirui Zhao et al. (+4 more)
Large
language models increasingly serve as long-horizon assistants and
agents, where they must both accumulate information across interactions and make the relevant parts available when later requests depend on them. Existing compact online
memories typically use a single persistent state both to accumulate history and to serve readout, so what the
memory stores cannot be controlled separately from what it exposes to the current computation. We propose LSTMem, an LSTM-inspired online
memory that instead equips each layer of a frozen LLM with two
matrix-valued states: a
cell state that accumulates history and a hidden state whose readouts correct the backbone's
attention. Input and forget gates control what the
cell stores, while an output gate separately controls what the
cell exposes through the hidden state. LSTMem further connects
memory across depth through forward hidden-state propagation and block-end feedback, and uses higher-layer reconstruction gradients to refine lower-layer
cell states before rebuilding hidden states from shallow to deep layers. Across
memory benchmarks on Qwen3-4B-Instruct, LSTMem consistently improves MemoryAgentBench, LoCoMo, and HotpotQA over the plain backbone. Comparisons further show that the LSTM-based
memory formulation outperforms an associative-
memory counterpart, while removing cross-layer hidden-
memory propagation degrades performance. These results demonstrate the benefits of separating
memory accumulation from
memory expression and organizing
memory hierarchically across model depth. The code is available at https://github.com/Longchentong/LSTMem.
Read the full abstract ▾
All authors Xianglong Shi, Ruijie Yang, Sirui Zhao, Shukang Yin, Zihao Bian, Tinghao Yi, Enhong Chen
General Relativity and Quantum Cosmology
Quasinormal modes of Hayward black holes from Schwarzschild to extremality: a comparative spectral analysis
Davide Batic and Denys Dutykh
We study scalar and electromagnetic quasinormal modes of the Hayward black hole and the axial spectrum of a specified Regge–Wheeler-type effective model, from the Schwarzschild limit to exact extremality. Multiprecision Chebyshev collocation reproduces most available low-lying oscillatory
benchmarks and supplies further overtone candidates. The
spin-2 equation is model dependent. It differs from the effective-source potential used in recent Hayward studies, and we quantify representative frequency differences. The discretisation also produces purely imaginary
eigenvalue sequences, whose physical classification requires special care because highly reproducible imaginary-axis ladders need not be poles of the continued Green function. Accordingly, the imaginary-axis entries reported here are retained as unclassified spectral candidates, not established QNMs. We quantify their finite-table gap dispersion and sensitivity to the number of included gaps. Their spacings near one quarter in $M\omega$, and the extremal product near $1/3$ with the throat radius, remain properties of the tabulated candidates, and they do not establish universal QNM spacing or throat-controlled damping.
Read the full abstract ▾