Strongly Correlated Electrons
Quantum quenches of scar states in the Affleck-Kennedy-Lieb-Tasaki model via Clifford augmented tensor network simulation
Azar C. Nakhl, Ben Harper, Muhammad Usman et al. (+1 more)
Clifford augmented methods have emerged as a powerful technique to simulate
quantum circuits and many-body systems by exploiting the stabiliser structure found in said systems, whilst additionally enabling one to readily quantify the non-stabilizerness (i.e.\ magic) present. In this work, we utilise a qudit Clifford augmented simulation method to probe the presence of magic in the Affleck-Kennedy-Lieb-Tasaki (AKLT) model, finding that its ground state and scar states are highly magical as quantified by their Stabiliser Rényi
Entropy (SRE). Nevertheless, owing to its low
entanglement structure, these states are still readily expressible as a Clifford augmented
Matrix Product State with low bond-dimension and moreover may be time-evolved using the Clifford augmented time-dependent variational principle (TDVP)
algorithm. We utilise Clifford augmented TDVP to study the string order dynamics in the scar states of the AKLT model after a quench within the Haldane phase to the anti-
ferromagnetic Heisenberg
Hamiltonian. We observe that there is a decay of string order following the quench, which is more pronounced with increasing bimagnon number. This finding highlights the sensitivity of string order in the AKLT scar states to
symmetry-preserving quenches that break the underlying restricted spectrum generating
algebra, even when the initial and final
Hamiltonians belong to the same topological phase.
Read the full abstract ▾
All authors Azar C. Nakhl, Ben Harper, Muhammad Usman, Thomas Quella
Machine Learning
Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics
Maoqi Liu, Junwei He, Bowen Zhang et al. (+5 more)
Rubric-based reinforcement learning (Rubric-RL) trains
language models where no verifier exists. A judge checks each criterion of a rubric, and the verdicts are aggregated into a reward, most often by a weighted sum. We show that this additive aggregation is the weak point. Under a sum, criteria compensate for one another: a policy that misses the one decision that matters can buy the points back with advice nobody asked for. On clinical consultation, such a policy scores higher and answers worse. Rubric coverage rises while appropriateness on held-out physician criteria falls below the untrained model. The medical criteria are not to blame. Grouped so that they must hold together, the same criteria, unchanged to the word, recover a third of the loss; shorter answers recover almost none. We therefore propose Protocol-level Rubrics (ProRubric), which keeps what the criteria ask for and changes how they are aggregated. It groups a checklist into a few protocol-level dimensions. A dimension counts only when all of its criteria hold and its failure clause does not fire. The grouping is done once, offline, and leaves the optimizer unchanged. ProRubric raises appropriateness by 10.8 points without losing coverage and has the best seven-
benchmark average at both scales. Reward validity is set not only by what a rubric verifies, but by how it aggregates. Code is available at https://github.com/Estrellajer/ProRubric
Read the full abstract ▾
All authors Maoqi Liu, Junwei He, Bowen Zhang, Feiran Li, Wentao Ma, Rongyi Lin, Shuhan Zhong, Quan Fang
Robotics
Coral Grow-out Robotic Assessment System (CGRAS): Scaling Coral Recruit Monitoring Through Robotics and Computer Vision
Dorian Tsai, Scarlett Raine, Emilio Olivastri et al. (+12 more)
Climate change is the largest threat to coral reefs, with increasing global impacts accelerating the need for scalable reef restoration technologies. Large-scale reef restoration depends on the mass production of corals, such as through coral aquaculture. Coral seeding with recruits grown in aquaculture facilities is a feasible restoration approach, but effective production requires consistent, high-frequency monitoring of tens of thousands of macroscopic (0.5-2mm diameter) recruits, making conventional manual assessment prohibitively labor-intensive. To address this monitoring bottleneck, we introduce the Coral Grow-out Robotic Assessment System (CGRAS) which combines robotic imaging and computer
vision to automate data acquisition, perform multi-species detection and counting of corals, and evaluate coral health. CGRAS automatically extracts coral growth, survival and spatial distribution metrics, with the aim of providing timely feedback to operators for optimizing production, grow-out and deployment workflow processes. We demonstrate CGRAS in a large aquaculture facility on standardized coral settlement tiles, reducing the time and labor costs by a factor of 9.6 as compared to manual monitoring, whilst achieving 96.4% agreement for Acropora kenti corals relative to expert counts.
Read the full abstract ▾
All authors Dorian Tsai, Scarlett Raine, Emilio Olivastri, Riki Lamont, Andrew Lui, Timothy Morris, Joshua Esplin, Christopher A. Brunner, F. Mikaela Nordborg, Reginald Wardleworth, Garima Samvedi, Karen Jackel, Matthew Dunbabin, Tobias Fischer, Andrea Severati
Signal Processing
Blind Interference Suppression in IRS-Aided Wireless Systems: A Statistical Channel Ratio Estimation Approach
Tao Wang, Xiaohui Zhang, Hehe Ban et al. (+3 more)
This paper addresses the problem of suppressing non-cooperative interference in intelligent reflecting surface (IRS)-aided wireless links without any channel state information (CSI) or cooperation from the interferer. We propose a fully blind framework that relies solely on received signal power measurements. A key insight is that nulling the aggregate interference channel requires only the complex ratios between the IRS-reflected paths and the direct interference link, rather than absolute CSI. We develop a novel estimation
algorithm that obtains unbiased estimates of these channel ratios using only power samples collected under random IRS configurations. Theoretically, we prove that unbiased estimation is feasible when the number of discrete phase levels $K\geq 3$, and establish the Cramer-Rao lower bounds (CRLBs) for both the phase offset and amplitude ratio estimates, thus providing design guidance. Based on the estimated ratios, we propose two low-complexity IRS phase optimization
algorithms: a one-shot greedy method and an iterative variant that mitigates error propagation from weakly reflecting elements. Simulations demonstrate that the proposed schemes can suppress strong interference to within a few dB of the interference-free upper bound, offering a practical, CSI-free solution for robust wireless communications in contested spectral environments.
Read the full abstract ▾
All authors Tao Wang, Xiaohui Zhang, Hehe Ban, Pengfei Lv, Yiwei Guo, Ming Yi
Number Theory
Monogenity of the irreducible factors of $(k,t)$-Fibonacci and $(k,t)$-Lucas polynomials
Michail Karatarakis and Sumandeep Kaur
In this article, we study the monogenity of the number field generated by a root of an irreducible factor of the generalized $(k,t)$-Fibonacci and $(k,t)$-Lucas
polynomial sequence. We use Dickson
polynomials and their connection with cyclotomic fields to prove our main results. As special cases, we recover the monogenity results for the classical Fibonacci and Lucas
polynomials and obtain corresponding monogenity results for Chebyshev
polynomials of the first and second kinds.
Read the full abstract ▾
Materials Science
Defect Dynamics and Isomer Quenching in $^{229}$Th:CaF$_2$ Crystals
Ming Guan, Michael Bartokos, Kjeld Beeks et al. (+16 more)
Thorium-229 doped calcium fluoride ($^{229}$Th:CaF$_2$) is a leading candidate for a solid-state
nuclear clock, owing to the unusually low energy of the $^{229}$Th isomer transition at 8.3557\,eV and the very high banggap of CaF$_2$ around 12\,eV. Here we report a temperature-dependent study of $^{229}$Th:CaF$_2$ luminescence, covering radioluminescence, photoluminescence, thermoluminescence, and afterglow. Using synchrotron x-ray excitation together with a cryo-
vacuum system, we resolve the thermal quenching of the self-trapped
exciton (STE) scintillation into its triplet and singlet components, identify eleven thermoluminescence glow peaks together with their associated emission bands, and separate the temperature-dependent components that make up the long-lived afterglow. The trap depths obtained in this way are closely correlated with the temperature dependence of x-ray-induced isomer quenching, including the enhanced isomer yield near \SI{-60}{\celsius} and the minimum near \SI{-80}{\celsius}, supporting the carrier-trapping picture in which trapped carriers are unavailable to quench the
nucleus. Taken together, these results connect the optical and
nuclear observables of $^{229}$Th:CaF$_2$ through a unified picture of electron defects and Th-related traps, while providing a comprehensive characterization of the material relevant to solid-state
nuclear clock operation.
Read the full abstract ▾
All authors Ming Guan, Michael Bartokos, Kjeld Beeks, Takahiro Hiraki, Shinji Kitao, Takahiko Masuda, Nobumoto Nagasawa, Koichi Okai, Ryoichiro Ogake, Martin Pimon, Noboru Sasao, Fabian Schaden, Thorsten Schumm, Makoto Seto, Kenji Tamasaku, Sayuri Takatori, Yoshitaka Yoda, Akihiro Yoshimi, Koji Yoshimura
Statistical Computation
Learn-Then-Differentiate Gradient Estimation
Nifei Lin, Qingkai Zhang and L. Jeff Hong
Learn-then-differentiate (LTD) estimates gradients by fitting a model to simulation outputs and differentiating it. We develop a unified framework explaining what LTD differentiates and how accurately it estimates gradients. For models with a weighted representation, LTD differentiates a learned representation of the underlying probability measure. We then show how accuracy guarantees for fitted models translate into guarantees for gradients and higher-order derivatives, with rates approaching the standard Monte Carlo rate under suitable smoothness conditions. The framework recovers established results for kernel regression, local
polynomial regression, and kernel ridge regression, and yields further guarantees for multiple kernel learning and smooth neural networks. These results provide a common foundation for understanding and analyzing LTD across learning methods.
Read the full abstract ▾
Numerical Analysis
A fully convergent fixed-point fast sweeping method with the WENO-JS local solver for steady state of hyperbolic conservation laws
Liang Li, Jun Zhu and Yong-Tao Zhang
The fixed-point fast sweeping methods with weighted essentially non-oscillatory (WENO) local solvers are a class of efficient and high-order accuracy numerical methods for solving steady-state solutions of
hyperbolic conservation laws. However, with the classical WENO-JS local solver, the iteration residue of high-order fixed-point fast sweeping scheme often has difficulty to settle down to round-off errors. To achieve the full convergence in a fast sweeping method, the fixed-point fast sweeping methods with non-traditional WENO local solvers based on unequal-sized substencils were designed. However, the WENO schemes based on unequal-sized stencils are more complex and in general more expensive in computational costs than the classical WENO-JS schemes. In this paper, we go back to the classical WENO-JS local solver and develop a new fully convergent fifth-order fixed-point fast sweeping method for solving steady-state problems of
hyperbolic conservation laws. Based on recent studies on the nonlinear weighting process around discontinuities of solution, we apply the technique of frozen weights for avoiding unnecessary adjustment of nonlinear weights in the WENO-JS local solver, which freezes the nonlinear weights once the residual sequence in the fast sweeping iterations has stabilized. Different from the existing work on frozen weights, we design a simple and robust approach to judge stabilization of iteration residues and determine the iteration step when the nonlinear weights are frozen in the fast sweeping method. Extensive numerical experiments on a wide range of challenging two-dimensional steady-state problems demonstrate that, unlike the previous fast sweeping method with the fifth-order WENO-JS local solver, the proposed new scheme consistently drives the iteration residues to round-off errors and achieves the full convergence.
Read the full abstract ▾
Machine Learning
scTrilemma: Balancing Identity, Invariance, and Fidelity in Single-Cell Representation Learning
Yunhak Oh, Yoonho Lee, Junseok Lee et al. (+9 more)
Single-
cell RNA-seq representation learning is fundamentally label-free:
cell identities, states, and contexts are not fixed training targets, so what constitutes signal or nuisance is analysis-dependent. A single representation must therefore preserve biological identity and state, remain robust to nuisance context, and retain the gene-level variation needed for expression analysis, three demands we call the representation trilemma. To tackle this problem, we introduce scTrilemma, a latent-bottleneck VAE that routes expression-derived variation to the
embedding, the decoder, or the prior rather than forcing all of it through one
embedding. It gates gene
tokens by expression, routes the
cell representation through the decoder, and conditions the prior on unlabeled pseudo-bulk context, under a single reconstruction objective and without target annotations or auxiliary representation losses. In release-based zero-shot evaluation on successive CZ CELLxGENE Census releases, scTrilemma leads all three demands at once and preserves biological-state, differential-expression, and pathway structure across multiple disease settings. Latent interventions further show that context can be removed at almost no cost to the other demands, leaving identity against fidelity as the remaining tension. Code is publicly available at https://github.com/yunhak0/scTrilemma.
Read the full abstract ▾
All authors Yunhak Oh, Yoonho Lee, Junseok Lee, Namkyeong Lee, Sang-Yeon Hwang, Yinhua Piao, Hyomin Kim, Seonghwan Kim, Jaechang Lim, Woo Youn Kim, Sungsoo Ahn, Chanyoung Park
Computer Vision
FrameMorrow: Future-guided Frame Selection with Prospective Tokens for Long-Horizon Video Generation
Bo Yin, Xiaobin Hu, Jiaqi Zhao et al. (+1 more)
Long-horizon
video generation requires models to effectively leverage an increasingly long generation history. As the generated history grows, retaining all previous content becomes increasingly expensive and redundant, making effective historical selection essential. Existing approaches often determine historical relevance based on the current content. However, information relevant to the present is not necessarily useful for future generation, while seemingly less relevant history may become important later. Our key insight is that historical information should be selected according to its relevance to future information needs. Capturing these needs does not require generating the full future; instead, a compact representation of what becomes important next is sufficient to guide historical selection. Building on this insight, we propose FrameMorrow, a prospective frame selector that predicts a small set of prospective
tokens representing future information needs and uses them to identify relevant information from history. FrameMorrow selects explicit historical frames rather than model-specific internal states, enabling plug-and-play integration across diverse generators, including closed-source models, with little additional inference cost. We evaluate FrameMorrow across five
benchmarks and 11 generative models spanning long-
video generation, interactive generation, and action-conditioned world models. Extensive experiments demonstrate consistent improvements in long-range consistency, visual quality, and action
alignment across diverse generation settings.
Read the full abstract ▾
All authors Bo Yin, Xiaobin Hu, Jiaqi Zhao, Shuicheng Yan
Differential Geometry
Green-Function Monotonicity under Positive Bakry-Émery Curvature
Yu-Zhao Wang and Zeng-Ting Wu
Inspired by the Green-function monotonicity formulas of Colding, Minicozzi, and Manea, we establish weighted monotonicity formulas on closed weighted Riemannian
manifolds satisfying ${\rm Ric}_f^m \ge (m-1)kg$, where $m>n\ge 3$ and $k>0$. Our results extend Manea's
elliptic monotonicity framework to the setting of finite-dimensional Bakry-Émery
curvature and provide a positive-
curvature counterpart to the weighted formulas of Song-Wei-Wu. Starting from the Green function of $-\Delta_f+m(m-2)k/4$, we derive an exact weighted Bochner identity and prove three monotonicity formulas, together with a one-parameter family for $\beta\ge (m-2)/(m-1)$, under explicit pole-integrability conditions. We also treat the case of negative effective dimension $m<0$, for which the nonnegative square decomposition remains valid, while the change in the behavior near the Green pole leads to reversed area and volume monotonicity on finite sublevel sets. We conclude with comparisons and questions concerning
entropy monotonicity formulas and positive-
curvature RCD spaces.
Read the full abstract ▾
High Energy Astrophysical Phenomena
Pulsar glitches triggered by quakes when stress reaches a threshold
Zixuan Zhou, Ruipeng Lu, Weiyang Wang et al. (+2 more)
The starquake model of
pulsar glitches is revisited in light of accumulated observational data. A statistical comparison between the time-predictable and the magnitude-predictable models, borrowed from seismology and grounded in the classical spring-slider framework, is performed against a sample of 245 glitch events from 9
pulsars. It is found that the data supports the time-predictable model, i.e., a quake resulting in a glitch could be triggered when the developed stress reaches a critical threshold, $\tau_{\rm c}$, while the remaining elastic energy (correspondingly the residual stress, $\tau_{\rm r}$) after quakes is not fixed. The critical stress could be the order of $\tau_{\rm c}\simeq (10^{34}-10^{35})\ {\rm dyn\,cm^{-2}}$ if the residual stress after a quake with maximum stress drops could be considered negligible (i.e., $\tau_{\rm r}\simeq 0$), to explain the glitch amplitudes and the intervals. This $\tau_{\rm c}$-value is the order of the shear modulus of strangeon
stars, but would be several orders higher than that of conventional
neutron stars.
Read the full abstract ▾
All authors Zixuan Zhou, Ruipeng Lu, Weiyang Wang, Han Yue, Renxin Xu
Quantum Physics
Joint measurability of coplanar POVMs
Taro Yoshino, Koichi Tojo, Kei Hagihara et al. (+2 more)
An unbiased
qubit positive operator-valued measure (POVM) can be uniquely expressed in terms of the Bloch vector. We prove that for POVMs with coplanar Bloch vectors, they are jointly measurable if and only if the perimeter of the convex closure of Bloch vectors is less than or equal to 4. Moreover, as a corollary, we give an affirmative answer to the
conjecture suggested by [Andrejic-Kunjwal 2020] on the existence of a joint device for POVMs whose Bloch vectors lie on the same circle.
Read the full abstract ▾
All authors Taro Yoshino, Koichi Tojo, Kei Hagihara, Akinori Tanaka, Takuki Tomita
Quantum Physics
Average-and Last-Iterate Lower Bounds for Optimistic Matrix Mirror-Prox in Quantum Zero-Sum Games
Yiheng Su, Emmanouil-Vasileios Vlatakis-Gkaragkounis and Pucheng Xiong
Optimistic
matrix mirror-prox (OMMP) computes $\epsilon$-approximate Nash
equilibria in
quantum zero-sum games with an $O(1/\varepsilon)$ average-iterate guarantee [arXiv:2311.10859]. We investigate whether this dependence on accuracy is tight and whether
geometric last-iterate convergence can be guaranteed. We study these questions through explicit games with one
qubit per player. First, we prove an $\Omega(1/\varepsilon)$ lower bound for the uniform-average output that includes the maximally mixed initial state, independently of the regularizer and step size. Second, we construct a fixed game on which optimistic gradient descent-ascent (OGDA), initialized at the maximally mixed state, has last-iterate Frobenius distance to
equilibrium $\Theta(1/t)$ and duality gap $\Theta(1/t^3)$ for every sufficiently small fixed step size. A separate fixed game exhibits arbitrarily long delays in reducing the initial error by a constant factor across a family of initial states. Finally, we give a fixed game with a unique, strictly complementary
equilibrium on which optimistic
matrix multiplicative weights updates (OMMWU) converge only polynomially from the maximally mixed state for every fixed positive step size. The last-iterate Frobenius distance and
quantum relative
entropy from the
equilibrium to the iterates decay as $\Theta(1/t)$, while the duality gap decays as $\Theta(1/t^2)$.
Read the full abstract ▾
Data Structures and Algorithms
Optimal VC Dimension of Contrastive Learning with Margin
Dionysis Arvanitakis, Vaggos Chatziafratis, Yiyuan Luo et al. (+1 more)
Contrastive learning is a successful paradigm for learning $d$-dimensional
geometric representations from a collection of “anchor–positive–negative” triplets $(i,j^{+},k^{-})$, indicating that “item $i$ is closer to $j$ than to $k$.” Despite its success, understanding why contrastive learning leads to representations of high generalization quality—beyond the often pessimistic predictions from PAC-learning—remains a central question. Recently, \citet*{alon2024optimal} proved that, for PAC-learning $d$-dimensional Euclidean representations of $n$-point
datasets, $\Theta(\min(nd, n^2))$ triplets are necessary and sufficient, while they posed as an open question whether their VC dimension bounds for the more realistic setting of contrastive learning with a margin can be improved. For a margin parameter $\alpha >0$, a triplet $(i,j^{+},k^{-})_{\alpha}$ is satisfied by the
embedding $\phi:[n]\rightarrow \mathbb{R}^{d}$, if $\|\phi(i)-\phi(k)\|_2>(1+\alpha)\cdot\|\phi(i)-\phi(j)\|_2$. In this work, we resolve their question by proving that the VC dimension of contrastive learning under any margin $\alpha\in(0,1)$ is in fact $O(n/\alpha^2)$, improving on the previous bound of $O(n\log(n)/\alpha^2)$. We also establish that the bounds are optimal up to constant factors, by providing a matching lower bound of $\Omega(\frac{n}{\alpha^2})$ (the previously known lower bound was $\Omega(\frac{n}{\alpha})$), for $\alpha\geq \max(n^{-1/2},d^{-1/2})$.
Read the full abstract ▾
All authors Dionysis Arvanitakis, Vaggos Chatziafratis, Yiyuan Luo, Konstantin Makarychev
Machine Learning
ReSCENE: Server-Side Replay for Structural Mitigation of Catastrophic Forgetting in Federated Continual Learning
Sungmin Kang, Zhengzhong Tu and Sunwoo Lee
Federated continual learning must integrate new tasks over time without losing earlier-task knowledge. Most existing methods attach an anti-forgetting mechanism to the client-trained, server-aggregated loop of
federated learning, which holds back new learning to preserve earlier knowledge and burdens resource-constrained clients. We propose ReSCENE, which structurally mitigates catastrophic forgetting by having each client upload a small condensed surrogate of its local data while the server keeps the surrogates of past tasks and trains the global model on them together with the current task surrogates. For efficient server
memory, we introduce temporal herding, which selects the more recent surrogates from the pool accumulated over a task into a compressed buffer. Our study provides a theoretical analysis showing that this buffer can represent the original task data more closely than full accumulation of all surrogates. Across CIFAR-10, CIFAR-100, and TinyImageNet, ReSCENE achieves the strongest accuracy over seven baselines, by up to $31.1$ points of average accuracy, while requiring as little as $0.11\times$ of the client computation and up to $179\times$ less upload than the model-update baselines. ReSCENE further demonstrates its effectiveness when scaled to larger client populations and larger models while remaining efficient, which makes it a practical method for
federated continual learning.
Read the full abstract ▾
Computation and Language
Scaling Parameter and Context in Attention: Native Sparse Attention from Mixture-of-Head
Zizhuo Fu, Runsheng Wang and Meng Li
Scaling
attention parameters can improve
language model quality, but retaining full
token histories makes additional heads costly at long contexts. Furthermore, since
attention retrieves and combines contextual information, parameter scaling should also support longer contexts. We therefore ask whether
attention parameter scaling can directly enable efficient and effective context scaling. We introduce NAMOH, an architecture-native sparse
attention mechanism that activates $K$ of $H$ heads per
token. Each head retains only its assigned
tokens and performs causal
attention within this subsequence. Head selection thus jointly determines active parameters and available context without scanning the full history. Under balanced assignments, increasing $H$ at fixed $K$ shortens head histories and reduces per-
token key-value (KV) access without increasing total KV storage. We further support head-relative rotary position
embeddings to shorten positional spans within routed subsequences, aiming to mitigate position-induced
attention noise. Experiments show that NAMOH can outperform fully activated models with the same total parameters, while enabling more efficient long-context inference than smaller dense models with matched active parameter counts. It remains compatible with GQA and existing sparse
attention mechanisms. We hope this work offers a new path for scaling
attention, with parameter scaling directly enabling context scaling.
Read the full abstract ▾
Computation and Language
Forging LLM Authorship Fingerprints with Targeted Rewriting
Haohan Yuan, Simin Chen, Xi Niu et al. (+3 more)
Model-attribution classifiers can often identify which
language model produced a text, making model-specific writing patterns a signal of provenance. Accurate attribution on unmodified text, however, does not show whether the prediction still identifies the original source after deliberate rewriting. We formulate this problem as targeted fingerprint transfer: rewriting one model's output so that attribution classifiers assign it to a chosen target model. We study summarization, where different models receive the same document and express the same underlying content, providing a controlled setting for conditional generation. We introduce ForgePrint, a search-then-distil framework that first searches for rewrites that move attribution toward a target fingerprint, then distils the selected rewrites into a one-pass 4B Student model. On CNN/DM, the Student reaches 70.2% target success rate, outperforming both its Teacher (54.1%) and the strongest of six published rewriting baselines (39.3%), against held-out classifiers that are never queried by the attack. It also reaches 68.3% target success when transferring summaries from an open model toward chosen commercial models. These results show that fingerprint detectability should not be conflated with source authenticity, and that text-only attribution can provide misleading evidence of model identity under targeted rewriting, even when it is accurate on unmodified text.
Read the full abstract ▾
All authors Haohan Yuan, Simin Chen, Xi Niu, Hanqing Guo, Depeng Xu, Haopeng Zhang
Machine Learning
SparLeak: Privacy Leakage from Sparse Attention in LLM Inference on Shared GPUs
Fahao Chen, Linkang Du, Jinhao Zhou et al. (+2 more)
Sparse
attention is widely used to accelerate long-context inference in modern large
language models (LLMs), but its input-dependent execution behavior introduces previously unexplored
privacy risks. We identify a new GPU micro-architectural side channel, termed Sparsity-Induced
Memory Access (SIMA), which arises from secret-dependent key-value cache access patterns induced by sparse
attention. Based on this observation, we present SparLeak, a phase-aware side-channel attack that extracts SIMA traces during LLM inference and enables two practical
privacy extractions: query attribute inference from prefill-phase traces and autoregressive response reconstruction from decoding-phase traces. By reconstructing approximate
token-level sparsity profiles from page-level observations and applying profiling-based learning, SparLeak accurately recovers sensitive information, including user-query attributes and private LLM response content. Extensive evaluation across three LLM architectures, three sparse
attention mechanisms, and three
privacy-sensitive
datasets shows that SparLeak achieves average attack success rates of 90.9% for attribute inference and 87.3% for response reconstruction under real-world LLM serving settings, highlighting the significance to account for SIMA leakage when deploying sparse-
attention-based LLM systems. We provide anonymized SIMA traces, trained attack models, evaluation scripts, and documentation as artifacts at https://anonymous.4open.science/r/Janus_artifacts/.
Read the full abstract ▾
All authors Fahao Chen, Linkang Du, Jinhao Zhou, Peng Li, Zhou Su
Artificial Intelligence
Diversity Combining for Multi-Path LLM Reasoning
Guangsheng Yu, Litianyi Zhang, Qin Wang et al. (+5 more)
Multi-path
reasoning methods such as self-consistency (SC) sample $K$
reasoning paths and choose the most frequent answer. However, their gains quickly plateau as $K$ increases, and locating this saturation point currently requires fitting a scaling model to accuracies measured at several values of $K$ or stopping online per query. We formalize multi-path LLM
reasoning as a diversity combining problem from wireless communications: each path is a noisy channel observation, and the pairwise correlation of path correctness caps the design-effect effective sample size of the vote at a finite ceiling. Generalized least squares (GLS) analysis shows that, for paths with equal gains, the optimal linear combiner of latent
embeddings is uniform exactly when every row of the path covariance has the same sum, as under exchangeability, supporting majority vote as the natural default in standard SC while leaving room for weighting or
pruning under heterogeneous
prompt-template branches. Across 5 models and 12
benchmarks,
prompt-template diversity reduces path correlation in $55$ of $60$ model-
benchmark cells, with a mean relative reduction of $16\%$ and reductions in every task domain. We derive an Adaptive-K rule that uses a four-path pilot to select $K^*$, retaining $96$–$103\%$ of MV@$K{=}32$ accuracy across Math, QA, and NLU.
Read the full abstract ▾
All authors Guangsheng Yu, Litianyi Zhang, Qin Wang, Xu Wang, Mingyuan Li, Shaoxiong Ji, Ren Ping Liu, Massimo Piccardi
Combinatorics
Upper bounds for ordered Ramsey numbers of forests and bounded-degree graphs
Lior Gishboliner and Xiangyu Li
We prove the following two upper bounds for ordered Ramsey numbers: (1) Every ordered forest $F$ on $n$ vertices satisfies $R_{<}(F,F)=O(n^{1+\lceil\log\chi_{<}(F)\rceil})$. This in particular answers a question of Geneson, Holmes, Liu, Neidinger, Pehova and Wass. (2) There is a function $f$ such that, for every fixed ordered graph $H$ with maximum degree at most $\Delta$ and interval chromatic number at most $k$, it holds that $R_{<}(H,K_n)=O_H(n^{f(\Delta,k)})$.
Read the full abstract ▾
Artificial Intelligence
More Choices, Fewer Decisions: Ordinal-Scale Bias in JEV-like Direct-Decision Models
Tianxiang Gao, Jinzhe Li, Zhiyuan Li et al. (+2 more)
Direct-decision models enable efficient classification and evaluation by selecting from predefined candidates, but their reliability depends on more than prediction accuracy. Across 40
datasets and nine models from the JEV, KEV, and StartLux Decision families, we identify a systematic tendency to underuse ordinal rating scales. We call this phenomenon ordinal scale-utilization bias: final decisions concentrate on fewer effective labels than the gold distribution. The bias is substantially weaker on tasks with unordered semantic categories. Evaluating the same items on finer ordinal scales strengthens compression despite controls for gold imbalance and candidate position. Randomizing candidate order reduces but does not eliminate compression, while widening ordinal gaps at fixed candidate count can improve accuracy without restoring utilization. For most models, concentrated decisions coexist with broad average candidate probabilities. Greater model size improves accuracy without consistently restoring scale utilization. Together, these findings characterize compression as strongly associated with ordinal structure and intensified by scale refinement, rather than fully explained by presentation position or local spacing. This bias can be effectively mitigated through targeted training, making scale utilization an actionable reliability target alongside accuracy. The code and data are available at https://anonymous.4open.science/r/jev_ordinal_scale_bia/.
Read the full abstract ▾
All authors Tianxiang Gao, Jinzhe Li, Zhiyuan Li, Yi Chang, Yuan Wu
Optics
Experimental Demonstration of Chiral Bound States in the Continuum Enabled by Time-reversal Symmetry Breaking
Hao-Chang Mo, Wen-Jin Zhang, Ze-Yu Wu et al. (+3 more)
Bound states in the continuum (BICs) provide a powerful route to optical resonances with vanishing radiation loss and enhanced light-matter interactions. Of particular interest are
chiral BICs, whose resonant states exhibit intrinsic circular polarization and
spin-selective radiation. Magneto-optical photonic
crystals have recently been predicted to host such states through time-reversal-
symmetry breaking. Here, we experimentally demonstrate
chiral BICs induced by time-reversal
symmetry breaking in a magneto-optical photonic
crystal, where an out-of-plane
magnetic field lifts a degenerate BIC pair into two nondegenerate resonances with opposite circular polarizations. Numerical simulations reveal the associated
chiral phase
vortices and strong
spin-selective radiation, while microwave measurements confirm the predicted behavior through angle-resolved transmission spectra and pronounced broadband circular dichroism. Our results establish an experimental platform for exploring
chiral BIC physics in magneto-optical photonic structures, opening new opportunities for
spin-selective photonics.
Read the full abstract ▾
All authors Hao-Chang Mo, Wen-Jin Zhang, Ze-Yu Wu, Xin-Tao He, Xiao-Dong Chen, Jian-Wen Dong
Mesoscale and Nanoscale Physics
Optical observation of interlayer spin correlation
Akiyoshi Park, Pranshoo Upadhyay, Andrey Grankin et al. (+16 more)
Spin correlations govern numerous collective behaviors of
quantum materials, underpinning exotic phenomena such as unconventional superconductivity and topological magnetism. In layered materials, the interlayer
spin correlation is important because it characterizes the
magnetic ground state and determines
spin transport across the interface. Yet interlayer
spin correlations have remained hard to measure directly, leaving one of the most basic quantities of two-dimensional magnetism out of experimental reach. Here we provide the first direct optical probe of interlayer
spin correlations, using two-
magnon Raman scattering in the van der Waals antiferromagnet (AFM) CrSBr, in remarkable agreement with a microscopic
spin-wave model without any fitting parameter. Moreover, the two-
magnon channel switches on only in the AFM state and vanishes when a
magnetic field takes the
crystal to a
ferromagnetic state. We furthermore establish an
exciton-mediated variant of the technique, where tuning the
laser near the
exciton resonance enhances the signal roughly tenfold due to the
exciton's large oscillator strength.
Magnon-pair Raman spectroscopy thus opens a direct optical window into interlayer
spin correlations in van der Waals magnets, extendable to twisted
bilayers and proximity-coupled
heterostructures.
Read the full abstract ▾
All authors Akiyoshi Park, Pranshoo Upadhyay, Andrey Grankin, Emil Viñas Boström, Mahdi Ghafariasl, Hassan Alnatah, Masoud Mohammadi-Arzanagh, Gautam Nambiar, Beini Gao, Alireza Alvandi, Sakthi Rajmano Madhan Kumar, Ghadah Alshalan, Isaac Sherwood, Mahmoud Jalali Mehrabad, You Zhou, Arun Ramanathan, Xavier Roy, Angel Rubio, Mohammad Hafezi
Robotics
PhaseSync-Exo: Human Clock Anchored Reference Adaptation for Dynamic Gait Tracking
Kaijie Qi, Yuehan Wang, Kaiming Xu et al. (+2 more)
Human-aware exoskeleton walking requires reconstructing gait, tracking diverse motions under dynamic constraints, and preserving human timing. We present PhaseSync-Exo, which combines two-IMU CNN-
Transformer reconstruction, factorized amplitude-cadence retargeting with curriculum-trained recurrent control, and a human-clock-anchored adapter (HCA). HCA combines human-clock attraction with
robot-relative feedback to adjust reference rate while preserving forward progression and continuity. The reconstruction module achieves a mean absolute error (MAE) of 3.24 degrees on a held-out recording, and the frozen tracking policy completes 356 of 357 amplitude-frequency trials. Two complementary dynamic comparisons isolate HCA's timing benefit without retraining. Against
robot-relative correction, HCA reduces human-clock phase MAE from 95.55 degrees to 10.99 degrees, limiting reference drift. When human and delivered phases initially differ, it reduces phase MAE from 72.02 degrees to 13.90 degrees versus fixed-clock continuation. Compared with immediate phase reset, HCA reduces transient reference-tracking hip RMSE by 22.7% without reference jumps. All 420 timing rollouts complete without falls. These simulation results support continuous phase acquisition and sustained
alignment to an independent human clock.
Read the full abstract ▾
All authors Kaijie Qi, Yuehan Wang, Kaiming Xu, Chong Li, Jiakuo Yu
Computer Vision
DecoMoE: Decoupling Visual Propagation and Expert Computation for Efficient Multimodal MoE Inference
Xudong Tan, Peng Ye, Ming Xie et al. (+4 more)
Multimodal mixture-of-experts (MoE) models combine sparse expert activation with visual-
language capabilities, yet their inference remains costly because long visual-
token sequences repeatedly incur
attention, routing, dispatch, and expert-MLP computation. Existing methods typically compress either the
token or expert dimension, leaving redundancy along the other. Our analysis reveals two complementary regularities: the depth required for visual propagation varies across inputs, while text-
token routing exhibits concentrated and recurrent expert-importance patterns. Based on these observations, we propose DecoMoE, a two-dimensional structured compression framework that decouples visual propagation from expert computation. The Sample-Adaptive Visual Boundary (SAVB) predicts an input-dependent visual-exit layer at which the visual-
token block is removed. The Routing-Calibrated Expert Prefix (RCEP) reorders experts offline using text-
token routed mass and, from this predicted exit layer onward, retains at each MoE layer the shortest contiguous prefix covering a target routed-mass fraction. We evaluate DecoMoE on Qwen3-VL-MoE and InternVL3.5-30B-A3B across six
benchmarks. On Qwen3-VL-MoE, DecoMoE retains 97.91% of dense-baseline performance while reducing computation from 27.06 to 16.73 TFLOPs and latency from 0.44 to 0.26 seconds, yielding a 1.69x speedup. Code will be available at https://github.com/ShawnTan86/DecoMoE.
Read the full abstract ▾
All authors Xudong Tan, Peng Ye, Ming Xie, Chenyu Huang, Yaoxin Yang, Jiayuan Fan, Tao Chen
Information Retrieval
SkillSeek: Revisiting Agent Skill Retrieval at Marketplace Scale
Guanqun Yang, Wenlong Zhang, Tian Shi et al. (+1 more)
Anthropic's
Agent Skills package reusable procedural know-how for an LLM
agent into SKILL.md directories, and open-source aggregations have grown past 230,000 skills, making selection rather than authoring the bottleneck. The standing answer in the literature outsources selection to the
agent itself: an LLM-mediated
retrieval loop that rewrites queries and refines candidates inside the
agent's decision loop, paying LLM
tokens on every task. We present SkillSeek, an open-source two-stage skill retriever built from the standard IR recipe (a BGE-base bi-encoder feeding a small cross-encoder, exposed over MCP). Across a $4 \times 11$ grid of pool, backbone, and method on the 89-task SkillsBench
benchmark, SkillSeek reaches observed parity with the LLM-mediated loop of Liu et al. at essentially no extra cost: plain bm25 alone records a pass rate at or above their refined loop on three of four settings, and a small cross-encoder covers the remaining difference on the fourth. A first-stage recall ceiling explains the pattern, and total per-trial spend drops from USD 51.30 to USD 27.54 (within fifty cents of the no-skill baseline). Under the SkillsBench tasks and OpenHands harness we tested, this positions the standard IR recipe as a strong default for
agent-skill
retrieval, with LLM-mediated alternatives a natural fit for cases where deterministic methods fall short.
Read the full abstract ▾
All authors Guanqun Yang, Wenlong Zhang, Tian Shi, Ping Wang
Algebraic Geometry
Divisors and harmonic morphisms on metric graphs of pseudocompact type
Benjamin Mascuch and Montserrat Teixidor i Bigas
A metric graph is of pseudocompact type if identifying parallel edges produces a tree. We give a constructive proof that, for such graphs, divisorial $d$-gonality is equivalent to the existence of a degree $d$ harmonic morphism to a tree. This mirrors the algebraic correspondence, for curves of compact type, between limit linear series of dimension one and admissible covers. We also deduce lifting results for positive-rank divisors, with a genus-preserving refinement when identifying parallel edges produces a path. Finally, we study the Brill-Noether theory in the path case.
Read the full abstract ▾
Computation and Language
BARRAC: Adaptation of an English Aspect-based Sentiment Analysis Approach for Classification Tasks in Arabic Dialects
Ali Almutairi, Gelareh Mohammadi, Imran Razzak et al. (+1 more)
With the rapid growth of Arabic NLP, several models,
datasets and
benchmarks have been reported. This paper asks whether approaches developed for majority
languages like English can be adapted to Arabic tasks. We adapt an English aspect-based sentiment analysis framework to Arabic classification tasks and present the adaptation as BARRAC: Brainstorming
Alignment and Replaced Representation learning for ArabiC tasks. BARRAC replaces consumer-review attribute pools with Arabic linguistic devices and markers for dialectal sentiment, sarcasm, and dialect identification, and replaces noisy self-training with two-stage training. Evaluated on five Arabic dialect
datasets, BARRAC achieves a mean macro-F1 of 63.93%, outperforming the best few-label SOTA by 3%, and outperforming GPT-4o on four out of five tasks. Error analysis provides insights into remaining challenges. These results demonstrate that adapting task-specific approaches is a promising direction for Arabic NLP alongside adapting models,
datasets and
benchmarks.
Read the full abstract ▾
All authors Ali Almutairi, Gelareh Mohammadi, Imran Razzak, Aditya Joshi
Computer Vision
Future Video Generation Better Aligns with the Human Visual Cortex than Observed Video
Chang-Bae Bang, Hyungjin Chung and Byung-Hoon Kim
Studying the
alignment between the internal representations of
vision models and the responses of the visual cortex to the same observed visual stimuli has enabled us to better understand human visual processing. However, studies so far have largely overlooked the fact that the human
brain not only processes observed visual stimuli, but also predicts upcoming stimuli based on what has been observed. Accordingly, we hypothesize that internal representations for generating future
video frames are better aligned with the predictive nature of human visual processing than representations of the observed
video itself. To this end, we compare the
alignment between human
video-watching fMRI responses in the visual cortex and the internal representations from two types of
video diffusion models, an autoregressive (AR) model and its non-AR base model. We first conduct a within-model analysis of the AR
video diffusion model and show that the representations for future
video generation align better with the visual cortex than the representations of the observed
video. We then compare the internal representations of the AR model with those of its non-AR base model and again show that the representations for future
video generation align better with the visual cortex than the representations for observed
video reconstruction by the base model. Specifically, the
alignment of observed
video reconstruction is concentrated in lower-order visual cortex, whereas that of future
video generation is concentrated in higher-order visual cortex. Finally, we show in a human behavioral experiment that humans prefer
videos generated by amplifying the contributions of individual layers that align better with the visual cortex.
Read the full abstract ▾