Peer Revenue Returns subject to peer review

Papers

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

All of arXiv

Monday 31 August

Algebraic Geometry

On the first-order deformations of seminormal affine toric varieties

We provide a general description of the module $T^1_X$ of first-order infinitesimal deformations of a not necessarily normal affine toric variety $X$ with a special emphasis on seminormal varieties satisfying Serre's $(S_2)$ condition. In the surface case we are able to give a very detailed and concrete picture which includes the dimension of each homogeneous component.
Read the full abstract

Soft Condensed Matter

Active Brownian Dynamics from a Hamiltonian Model: Transitioning from Equilibrium to Activity

We introduce a Hamiltonian Active Brownian Particle (HABP) model that connects equilibrium dynamics with the non-equilibrium, two-dimensional overdamped behavior of standard Active Brownian Particles (ABPs). In equilibrium, the system follows overdamped Langevin equations that strictly satisfy the fluctuation-dissipation theorem. Coupling translational and rotational degrees of freedom to separate heat baths ($T_{\theta} > T_{\textrm{tr}}$), drives the system out of equilibrium. In free space, matching the diffusion coefficients yields quantitative agreement with the ABP model in the limit $T_\theta/T_{\textrm{tr}}\to \infty$, whereas in a harmonic potential, this matching is recovered even at finite temperature ratios. Entropy production analysis demonstrates that in the ABP limit, all energy injected by the swim force dissipates into the translational bath, leaving the rotational bath as a zero-cost entropy source. These insights provide the first steps toward equilibrium-inspired descriptions of active matter, facilitating the development of new theoretical frameworks to study activity-induced fluctuations, transport, and phase behavior in living and non-living soft matter systems far from equilibrium.
Read the full abstract

Game Theory

Reaching Fairness by Reallocating Goods

Fair allocation of indivisible goods has largely been studied under the assumption that no prior allocation exists. Motivated by practical settings with pre-existing (and possibly unfair) allocations, we study how to achieve fairness through limited reallocations. Building on recent work on reformability/reallocations, we consider three fairness notions—envy-freeness (EF), envy-freeness up to one good (EF1), and envy-freeness up to any good (EFX)—and optimize the number of goods reallocated. We analyze both the classical and parameterized complexity of these problems, providing a comprehensive analysis across multiple fairness notions.
Read the full abstract

All authors Robert Bredereck, Eva Deltl, Tanmay Inamdar, Pallavi Jain, Pranjal Pandey, Bin Sun

Probability

Recurrence and capacity of stable branching random walks

We study the linear growth rate of the range of size-conditioned Branching Random Walks (BRW) when the offspring distribution $\mu$ is critical and attracted to an $\alpha$-stable law. This is done via the infinite invariant BRW introduced by Le Gall & Lin and a new criterion which relates this growth rate of the range to a notion of dimension of the underlying tree in a general way. Then, in the transient case (that is, when the range does grow linearly), we extend the notion of branching capacity to this $\alpha$-stable case. We show that it is still related to the asymptotic probability that a BRW (or its infinite version) reaches a distant set in $\mathbb Z^d$, and we estimate the $\alpha$-stable branching capacity of balls.
Read the full abstract

Data Structures and Algorithms

The (Parameterized) Complexity of Ordering a Graph While Avoiding a Forbidden Pattern

In this paper, we study the Pattern Avoidance problem of determining whether a given graph $G$ admits a linear vertex order which avoids a given pattern $P$, i.e., a vertex sequence with some forced and forbidden edges, on every suborder. Such patterns form a natural ordered counterpart to induced subgraphs in the order-invariant setting, and it is known that Pattern Avoidance captures a broad variety of graph problems including Bandwidth, Vertex Coloring, Queue Number, and extends to vertex-deletion problems such as Odd Cycle Transversal. We show that Pattern Avoidance is $\Sigma_2^{\textsf{P}}$-complete and furthermore remains intractable (in both the classical and parameterized sense) even under a variety of severe restrictions to both the pattern $P$ and the graph $G$. As our main contributions, we complement these lower bounds with the following tractability results, which provide a unifying framework for recognizing pattern-definable graph classes: - a fixed-parameter algorithm w.r.t. the vertex integrity of $G$ plus $|V(P)|$, - a fixed-parameter algorithm w.r.t. the neighborhood diversity of $G$ plus $|E(P)|$, and - a polynomial algorithm for Pattern Avoidance on forests for almost all constant-sized patterns.
Read the full abstract

All authors Thomas Depian, Simon D. Fink, Alexander Firbas, Robert Ganian, Martin Nöllenburg, Marie Diana Sieper

Instrumentation and Methods for Astrophysics

On-sky demonstration of a vector Zernike wavefront sensor in a cascaded adaptive optics system

To directly image and characterise Earth-like exoplanets, future high-contrast instruments will require adaptive-optics systems operating at increasingly high loop frequencies to reduce temporal errors. Increasing the loop frequency reduces the signal-to-noise ratio per wavefront-sensor frame, making highly sensitive wavefront sensors, such as the Zernike wavefront sensor (ZWFS), attractive candidates. However, the limited dynamic range of the classical ZWFS makes on-sky operation challenging. We therefore investigate whether a ZWFS can be used as a second-stage sensor in an on-sky cascaded adaptive-optics system. To this end, we added a second AO stage, called OZIRIIS, to the PAPYRUS platform at the Observatoire de Haute-Provence. OZIRIIS combines a vector Zernike wavefront sensor (v-ZWFS) with a 97-actuator deformable mirror operating at 400 Hz downstream of the pyramid-based first AO stage. Real-time control relied on a single ZWFS signal, while the full v-ZWFS was used a posteriori for non-linear reconstruction and telemetry analysis. The second-stage correction increased the measured Strehl ratio by up to 16 percentage points. Analysis of the telemetry using the full v-ZWFS to reconstruct residuals revealed optical-gain effects affecting the ZWFS at low Strehl ratio. The good agreement between on-sky measurements and numerical simulations further supports the calibration strategy based on synthetic reference signals and interaction matrices. These results demonstrate that Zernike wavefront sensing can be operated in closed loop on sky and support its use in future cascaded extreme adaptive-optics systems.
Read the full abstract

All authors M. Motte, V. Chambouleyron, R. Fétick, F. Oyarzun, M. A. Alagao, A. Striffling, E. Vinerskas, J.-F. Sauvage, C. T. Héritier, E. Muslimov, M. Cissé, A. Rahim, J. Kent Wallace, T. Wenger, B. Neichel, T. Fusco

Materials Science

Terahertz Control of Optical Second-Harmonic Generation in Displacive Ferroelectrics: Electronic Bloch-State Reconstruction

Optical second-harmonic generation (SHG) is a powerful probe of ferroelectric order, yet its microscopic origin and dynamical control are often understood primarily from symmetry considerations rather than from the underlying electronic processes. Here, we develop a microscopic theory of terahertz-controlled optical SHG in displacive ferroelectrics, establishing a direct connection between THz-driven polar lattice distortions and the resulting electronic nonlinear optical response. Starting from a complete Bloch-band representation, we show that an inversion-breaking lattice distortion reconstructs electronic Bloch wave functions, modifies optical dipole matrix elements, and activates nonlinear optical pathways that are forbidden in the centrosymmetric structure. We derive the second-order susceptibility in terms of the distortion-induced reconstruction of electronic states and optical transition matrix elements, and demonstrate that the electronic SHG susceptibility is linear in the polar distortion, $\chi^{(2)}({\bf Q})\propto{\bf Q}$, leading to an SHG intensity quadratic in the inversion-breaking order parameter. When the polar mode is coherently driven by a terahertz electric field, the resulting time-dependent lattice distortion dynamically reconstructs the electronic states and thereby modulates the optical SHG response, with $I_{2\omega}(t)\propto|{\bf Q}(t)|^2$ to leading order. This framework distinguishes the THz-driven lattice dynamics from the electronic interband processes responsible for optical SHG, which is particularly important in insulating ferroelectrics where low-energy carrier dynamics are absent. Our theory thus provides a microscopic bridge between nonequilibrium polar lattice dynamics and ultrafast electronic nonlinear optics.
Read the full abstract

All authors Hong-Kui Liu, Zi-Chen Qin, Jun-Song Wu, Yue Yuan

Computational Complexity

Upper and lower bounds on the OBDD-width of a special integer multiplication

We consider the Boolean function ${\rm SMul}_{n-1}^n(\boldsymbol{x},\boldsymbol{y})$, which computes the middle bit of the multiplication of two natural numbers represented as $n$-bit binary strings $\boldsymbol{x}$ and $\boldsymbol{y}$, drawn from a restricted domain. We investigate the width of OBDDs computing ${\rm SMul}_{n-1}^n$. We introduce a combinatorially defined function $s_*(n)$ and show that the width of such OBDDs is $\Theta(2^{s_*(n)})$.
Read the full abstract

Signal Processing

Event-Inference Reliability for Physical AI over Wireless Networks

Wireless-enabled physical artificial intelligence (physical AI) systems call for a shift from reliable data delivery to reliable inference of physical events. The relevant question is not only whether packets arrive, but whether the set of cues available at the decision node, i.e., the evidence, is sufficiently timely and informative to support reliable inference about the event. Accordingly, this paper develops a framework in which event-inference reliability (EIR) is determined jointly by cue informativeness, cue availability, and temporal admissibility. The latter is determined by the downstream task requirement and represented through the usefulness horizon. We define event-inference error ratio (EIER) as the normalised residual event uncertainty after incorporating admitted cues, and EIR as the corresponding normalised uncertainty reduction, both conditioned on decision-node context. We further distinguish the evidence-limited Bayes benchmark from operational performance of a particular inference engine and derive an entropy-based lower bound on the minimum achievable event error from the same decision-node information. The framework then enables an event-aware wireless design interface for cue prioritisation, cue-reliability allocation, and event-inference coverage characterisation. A multiclass indoor activity-inference study combining empirical cue likelihoods with wireless delivery instantiates the framework and demonstrates how it characterises EIR under finite usefulness horizons.
Read the full abstract

Computation and Language

MURANO: Design, Run, and Reproduce Mechanistic Interpretability Experiments as Composable Pipelines

This paper presents Murano, an open source framework for designing, running, and reproducing mechanistic interpretability studies of large language models, intended for researchers across disciplines. These studies often combine loading, recording, attribution, intervention, and evaluation, while existing libraries tend to focus on different parts of this workflow. As a result, researchers using several libraries may need to adapt outputs from one for use by another. To bridge this gap, Murano represents operations from these five areas as composable steps. Steps exchange named result artifacts and declare the inputs they require and the outputs they produce. A pipeline executes its steps in the order supplied, and Murano uses canonical addresses when component identities pass between operations. Murano builds on existing interpretability and machine learning libraries. We demonstrate Murano through two reproductions of established interpretability studies and one illustrative sparse autoencoder case study.
Read the full abstract

All authors Alireza Bayat Makou, Emirhan Böge, Phu Gia Hoang, Federico Tiblias, Jingcheng Niu, Subhabrata Dutta, Richard Eckart de Castilho, Iryna Gurevych

Computation and Language

SwarmBench: Can Large Language Models Act as Agent Swarm Orchestrators?

Large language model-based multi-agent systems are evolving from fixed interaction topologies toward dynamically orchestrated Agent Swarms. However, existing benchmarks are still largely based on single-agent or general-purpose agent tasks, making it difficult to systematically evaluate key orchestration capabilities. We propose SwarmBench, a benchmark that evaluates model performance from multiple perspectives, including accuracy, efficiency, cost, and process quality. Experimental results show that current models exhibit substantial differences in orchestration capability. These differences are reflected not only in final accuracy, efficiency, and cost, but also in the overall quality of the orchestration process itself. Based on these findings, we further propose SwarmExp, a simple yet effective method based on experience extraction and experience replay, which consistently improves the orchestration performance of large language models.
Read the full abstract

All authors Jinshan Gao, Zhuoran Jin, Tianyi Men, Kang Liu, Jun Zhao

Quantum Physics

Trade-off between Cooling-Step Count and Geometric Implementation Cost in Non-Markovian Algorithmic Cooling

Quantum cooling is important for reliable quantum computation but involves a trade-off between cooling performance and implementation resources. Although reservoir memory can improve particular aspects of cooling performance, the associated resource cost, particularly for circuit implementation, remains insufficiently understood. Here, we investigate how reservoir memory affects the trade-off between cooling performance quantified by the cooling-step count and geometric implementation cost in single-qubit heat-bath algorithmic cooling. Using a pseudomode mapping, we represent the non-Markovian damped Jaynes–Cummings dynamics by a repeated collision circuit and evaluate its geometric implementation cost. Using matrix-based simulations and an implementation on the ibm_kawasaki Heron r2 processor, we identify a trade-off: suppressing reservoir memory reduces the cooling-step count but generally increases the geometric protocol cost. Our work provides a resource-based perspective on reservoir engineering for algorithmic cooling.
Read the full abstract

Hardware Architecture

LLM-based Hardware Development with Hierarchical IRs and End-to-End Multi-Agent Workflow

Large language models (LLMs) are increasingly used in software development, but their use in complex hardware design remains limited. This gap stems from both the scarcity of public hardware training data and the fundamentally different methodologies used in hardware design. In particular, applying LLMs to hardware requires more than direct RTL generation: the model must understand module boundaries, inter-module connections, and verification requirements. In this paper, we present an LLM-based hardware development framework with hierarchical intermediate representations (IRs) and an end-to-end multi-agent workflow. The core idea is to provide an abstraction of hardware design to LLMs through two structured IRs: Architectural Sketch, which captures module topology and interconnection, and Operational Specification, which defines per-module functionality and interfaces. Our framework uses these IRs to decompose a complex design into sub-modules, specify the per-block functionality, and derive how each module should be tested and verified. We incorporate a multi-agent debug loop in the framework, allowing agents to get the error feedback and control the debug details such as the signals to be probed for simulation. We evaluate our framework on Verilog-Eval benchmark, achieving a pass@5 rate of 95.5%, which surpasses current state-of-the-art LLM generation frameworks. To better assess performance on complex, realistic designs, we introduce a new case study spanning applications from general-purpose processors to digital signal processing systems. Experimental results indicate that such complex designs exceed the capabilities of existing approaches, whereas our framework is the only one capable of producing functional end-to-end design. Our generated RTL follows all industry-standard design rules, is lint-clean, functionally correct and fully synthesizable.
Read the full abstract

All authors Chenyang Yin, Agasthi Haputhanthri, Aditya Anirudh Jonnalagadda, Zhenyu Bai, Yuanming Song, Saranyu Chattopadhyay, Mohammad Fadiheh, Tom Zelazny, Subhasish Mitra, Tulika Mitra

Analysis of PDEs

Higher Regularity of Time-Periodic Solutions to Nonautonomous Hyperbolic Problems: Away from Resonances

We study higher regularity and its relation to nonresonant behavior for time-periodic solutions of boundary value problems for one-dimensional linear and nonlinear nonautonomous first-order integro-differential strictly hyperbolic systems. The boundary conditions include integral operators and various types of boundary reflections. We prove that continuous and classical solutions have $C^k$-regularity, provided the coefficients are sufficiently smooth and a suitable number of nonresonance conditions is satisfied. In the linear case, these conditions involve the principal coefficients, the diagonal lower-order coefficients, and the boundary reflection coefficients. In the nonlinear case, they also depend on the nonlinearities and on the solution itself. For nonautonomous hyperbolic systems, higher regularity generally requires additional nonresonance conditions, whose number depends on the desired order of differentiability. These conditions are not only sufficient but, in general, also necessary, revealing a distinctive feature of nonautonomous hyperbolic PDEs. By contrast, in the autonomous case, a single nonresonance condition (if one is needed at all) suffices to obtain arbitrarily high regularity. We also identify a class of nonautonomous hyperbolic problems for which no nonresonance conditions are required. In this case, the higher regularity of solutions is determined solely by the regularity of the data. The main technical tool underlying the proofs is an abstract regularity principle formulated in the setting of vector spaces.
Read the full abstract

Computer Vision

InfraOcc: An Infrastructure Occupancy Benchmark with Static-to-Dynamic Reasoning

Fixed-viewpoint infrastructure sensors repeatedly observe the same traffic space, making roadside 3D occupancy structurally different from ego-vehicle perception: a near-persistent static scaffold is overlaid with sparse, short-lived dynamic events. Existing occupancy benchmarks and methods, however, are built around moving ego vehicles and neither measure nor exploit this structure, instead treating occupancy as flat one-shot voxel classification. We address this gap from both data and model perspectives. We build InfraOcc, to our knowledge, the first real-world infrastructure-side semantic occupancy benchmark, with dense voxel annotations for 290 multi-modal sequences in a fixed roadside frame, a static-dynamic decoupled annotation pipeline, unified camera-only, LiDAR-only, and multi-modal evaluation, and diagnostics for static and dynamic occupancy. InfraOcc shows that static infrastructure fills 97.3% of occupied voxels and persists across frames, whereas dynamic participants have a median occupied-frame ratio of only 1.8% per location, revealing a structural static-dynamic asymmetry beyond semantic long-tailedness. We further propose ProSD-Occ, which reformulates occupancy as progressive static-to-dynamic evidence reasoning: it explains persistent layout, exposes residual dynamic evidence under static-confidence guidance, and recomposes static, dynamic, and free-space evidence into a unified field. ProSD-Occ ranks first in overall, dynamic, static, and geometric occupancy on every track, e.g., a 23.5% relative camera-only dynamic-mIoU gain over the strongest baseline and 65.87 multi-modal overall mIoU, establishing fixed-viewpoint roadside occupancy as a distinct problem with its own reasoning paradigm. The benchmark and code will be publicly available at https://github.com/yanglei18/InfraOcc
Read the full abstract

All authors Lei Yang, Xiaokai Bai, Boqi Li, Chunmian Lin, Li Wang, Ziying Song, Jiahuan Zhang, Enhui Ma, Haibao Yu, Jiaqi Ma, Kaicheng Yu

Computer Vision

APT: Anchor-aligned Perturbations for Tamper Localization in Fully Regenerated Images

Proactive tamper localization embeds an imperceptible signal into an image prior to distribution, enabling pixel-level manipulation detection. Existing methods assume a spliced (SP) setting, where synthesized regions are composited onto the original background, leaving embedded signals intact. However, real-world diffusion-based inpainting operates in a fully regenerated (FR) setting, where the entire image undergoes denoising, disrupting background signals and rendering existing frameworks ineffective. We propose APT, a semi-fragile latent-space perturbation that embeds a dense, vector-wise localization signal. By aligning each spatial feature vector toward a fixed anchor direction, APT localizes tampering via the alignment disparity between synthesized foreground and anchor-aligned background features after inpainting. The proposed hard negative mining loss and noisy perturbation branch further enforce uniform alignment. Experiments on COCO demonstrate that APT achieves an FR IoU of 0.92, outperforming the strongest baseline (WAM, 0.84), while existing methods collapse to near-random performance (AUC 0.5), establishing APT as a practical forensic framework generalizable across tampering types unknown at test time.
Read the full abstract

All authors Suhyeon Ha, Woo Jae Kim, Joonsung Jeon, Sooel Son, Sung-eui Yoon

Populations and Evolution

Dimensionality-induced critical phase transition in stochastic Lotka-Volterra equation: From statistical averaging to systemic tipping point

By analyzing a stochastic Lotka-Volterra (LV) equation, we show that the underlying logic behind the diversity-stability debate in ecology can be framed as a dimensionality-induced critical phase transition, that is, a transition separating the regime dominated by statistical averaging and a systemic tipping point. This phase transition framework unifies the opposing ecological predictions in the diversity-stability debate and sets an intrinsic diversity ceiling for stochastic ecological communities.
Read the full abstract

All authors Xiu-deng Zheng, Cong Li, Hui Zhang, Hui-jie Qiao, Shao-peng Wang, Yi Tao

Machine Learning

Seasonality-Aware Hybrid Convolutional Transformer for Antarctic Sea Ice Concentration Forecasting

Antarctic sea ice concentration (SIC) forecasting is an important yet challenging task due to the coexistence of complex spatial structure, long-range temporal dependencies, and strong seasonal variability. Conventional convolution-based models are effective at capturing local spatial patterns, but often have limited ability to model long-term temporal evolution. To address these challenges, we build on a hybrid convolutional transformer forecasting framework for monthly Antarctic SIC forecasting. This framework combines convolutional encoding for spatial feature extraction with space-time factorised self-attention for SIC modelling. We further introduce two season-based mechanisms: a month-aware positional encoding that injects calendar-month information into the token representation, and a seasonal temporal bias that encourages attention to periodically related historical states. Experimental results show that the proposed framework achieves better performance than convolutional and recurrent baselines as well as ECMWF's physics-based dynamical model SEAS5 under both classification and regression metrics. Ablation studies further indicate that the seasonality-aware components provide consistent additional gains in both short- and long-horizon prediction. These results demonstrate the value of combining convolutional structures, attention mechanisms, and periodic prior information for Antarctic SIC forecasting.
Read the full abstract

All authors Danyang Li, John Taylor, Thang Bui, Quanling Deng

Computer Vision

Fine-Grained Multi Image Object Hallucination Benchmark

Multimodal Large Language Models (MLLMs) are increasingly deployed in multi-image scenarios requiring complex reasoning across visual contexts. However, current MLLMs remain fundamentally limited by object hallucination-generating plausible yet factually inconsistent descriptions about objects. Existing benchmarks, designed primarily for single-image settings or providing only high-level multi-image assessments, cannot systematically diagnose how visual complexity and reasoning demands trigger hallucination. To address this gap, we introduce MIOH, a fine-grained multi-image object hallucination benchmark that systematically evaluates object hallucination across four foundational tasks (existence, counting, attribute, position) through three multi-image reasoning patterns (comprehensive, comparative, selective) under three controlled adversarial pressures (visual context scale, perceptual difficulty, contextual bias). Through evaluation of 29 models, we reveal that even state-of-the-art systems like GPT-5 and Gemini-2.5-Pro exhibit distinct failure patterns across different reasoning patterns and tasks. Our evaluation reveals that hallucination stems not merely from perceptual failures but from integration-stage limitations when maintaining object representations across multiple images. MIOH provides a controlled framework for analyzing multi-image object hallucination and serves as a critical evaluation tool for developing more reliable multimodal AI systems.
Read the full abstract

All authors Joonki Min, Chaeyun Kim, Hyungwook Choi, Yejin Kim, Kihyun Kim, Yohan Jo, Joonseok Lee

Artificial Intelligence

PyKEEN-NSX: A Modular Framework for Static, Dynamic and Schema-Aware Negative Sampling in PyKEEN

Embedding methods have become popular due to their scalability on link prediction and/or triple classification tasks on Knowledge Graphs (KGs). Embedding models are trained relying on both positive and negative samples of triples. However, since KGs generally contain only positive assertions, negative samples are artificially generated through negative sampling strategies, ranging from simple random corruption to more sophisticated approaches that exploit structural, semantic, or embedding information. The design and implementation of advanced negative samplers remains challenging, as most popular Knowledge Graph Embedding (KGE) libraries provide support only for basic strategies and lack a unified framework for developing more advanced and customized solutions. To address this gap, we introduce PyKEEN-NSX, an extension of PyKEEN, the popular KGE framework, that provides a modular engineered abstraction for negative sampling. The proposed architecture separates the generation of candidate negative pools, conditioned on an explicit context, from the selection strategy, enabling the development and integration of static, schema-aware and dynamic approaches within a consistent framework. Based on this abstraction, we implement six negative samplers, while remaining fully compatible with existing PyKEEN workflows and pipelines. As a proof of concept, we study negative availability across four datasets, showing that constrained pools frequently fall below the requested number of negatives, so that the encoded criterion is to a large extent replaced by the random fallback that supplements them.
Read the full abstract

Spectral Theory

On the discrete spectrum of Dirac operators with Lorentz-scalar $\delta$-shell interactions supported on unbounded curves

We consider the massive Dirac operator (with positive mass) in the plane with an attractive Lorentz-scalar $\delta$-shell interaction of strength $\tau\in(-\infty,0)\setminus\{-2\}$ supported on a $C^\infty$-smooth curve $\Sigma\subset\mathbb{R}^2$ being a local deformation of the broken line. This singular interaction is defined by imposing a suitable transmission condition on the curve $\Sigma$ in the operator domain. Such a Dirac operator is self-adjoint and has a gap in the essential spectrum, whose size is explicit and depends on the mass and the interaction strength. We show that the number of discrete eigenvalues in the gap is finite. Under the assumption that one of the domains bounded by $\Sigma$ is convex, we prove that the corresponding Dirac operator has a non-empty discrete spectrum, provided that $\tau$ is either sufficiently small or sufficiently large in absolute value. The result holds for any perturbation of a broken line of any opening angle as described above, and this discrete spectrum is induced by the geometry, since for the same type of a singular interaction supported on the straight line the discrete spectrum is empty.
Read the full abstract

Artificial Intelligence

Geometry of Divergence: Tracking Hidden-State Trajectories for Adaptive Multi-Turn Reasoning

LLM agents need to sustain goal-consistent reasoning across long multi-turn interactions under strict resource constraints. However, as the multi-turn context accumulates, it can destabilize the underlying LLM's internal representation of task-relevant information from earlier turns, blurring the boundary between constructive reasoning and representation drift. We formulate multi-turn reasoning as a hidden-state trajectory of the underlying LLM that is characterized via two complementary signals: temporal curvature that captures the directional consistency of turn-to-turn updates, and variance slope which measures the expansion or contraction of the exploration space. Across four tasks and three underlying LLMs, we observed that these geometric signals distinguish between correct and incorrect episodes prior to completion. We further decompose each episode into three-action chains formed from four actions (Read, Write, Respond, Transfer) and show that separability is action-dependent, with different signals distinguishing various chain patterns. Our experiments demonstrate that trajectory geometry can identify critical turns in the reasoning process, increasing task success rates on $\tau$-Bench from 24.1% to 39.6% while reducing token cost by 11.2%.
Read the full abstract

All authors Jie Liang, Zhengxin Yu, Hamid Nasiri, Peter Garraghan

Computation and Language

Where Identity Lives: Localized, Retain-Free Identity Unlearning in Multimodal Large Language Models

Removing a specific individual's information from multimodal large language models (MLLMs) is often needed after deployment, but existing methods rely on a retain set, which is hardest to obtain at that point, and rebuilding it recreates the privacy exposure that unlearning aims to remove. Forgetting from the forget set alone instead damages the shared visual-language computation, harming perception. We cast retain-free unlearning as a localization problem: causal tracing, weight transplant, and Fisher overlap all point to early-to-mid decoder MLPs as the layers where identity information is stored and, unlike other module families, can be modified without substantially disrupting vision. We turn this into Pathway-Aware Visual-attribute Anchoring (PAVA), which confines updates to these layers and pairs a forget loss with a visual-attribute anchor that preserves image-grounded behavior by distilling the model's own pre-unlearning answers from the forget images alone. On MLLMU-Bench and ReMem, PAVA gives the strongest forget-retain trade-off among forget-set-only methods and remains competitive with retain-based baselines.
Read the full abstract

All authors Kangwook Ko, Jaehyuk Jang, Wonjun Lee, Hee-Seon Kim, Changick Kim

Cryptography and Security

Lie to Me: Finding Bugs in ZK DSL Toolchains with Adversarial Witness Injection

Zero-knowledge domain-specific language (ZK DSL) toolchains compile programs into constraint systems and generate witnesses for cryptographic proofs. Bugs in these toolchains can leave the enforced constraints weaker than the source-program semantics, admitting proofs for invalid executions. Such soundness bugs may remain invisible to valid-execution testing because all valid executions still behave correctly. We present Liezz, a testing framework that generates ZK DSL programs and exposes these bugs through adversarial witness injection. For each generated deterministic program, Liezz executes two public input assignments with different outputs and splices their witnesses, combining the input of one execution with the output of the other. The resulting witness is invalid by construction. A correct toolchain must reject it; acceptance exposes a soundness bug. Controlled divergence and multiple witness-splicing strategies preserve enough consistency to expose missing constraints. Liezz also generates parameterized standard-library calls to reach complex functionality. Liezz supports Circom, Corset, Gnark, and Noir. It finds 13 bugs, including seven with soundness impact. Several are reachable only through generated standard-library calls. Under the same testing budget, a valid-execution baseline does not expose any of the soundness failures revealed by accepted injected witnesses, showing that adversarial witness injection reaches failures missed by valid-execution testing.
Read the full abstract

All authors Sebastian Watzinger, Christoph Hochrainer, Valentin Wüstholz, Maria Christakis

Computation and Language

What It Costs to Compose, Rebuild, and Correct Precomputed Memory

Language models can answer from precomputed memory, a model's saved reading of a body of material, reused across requests instead of read again at each. This paper maps where that practice preserves correctness and the conditions under which it fails. Across experiments on Llama-3.1-8B-Instruct using both saved key-value caches and trained compressions of them, precomputed memory degrades when assembled from separately prepared parts, stays current only through rebuilds costing a large fraction of full preparation in our measurements, and ignores corrections served beside it conditional on phrasing. If precomputed memories can be served alongside one another, be cost-efficiently rebuilt, and be superseded by new information arriving in real-time, they can serve as a way to avoid re-feeding context to a model over repeated queries. The implication of our results for a deployed system that deals with a variety of queries is that precomputed memories are best rebuilt on the cadence at which new information changes what the memory was originally computed from. Both warm-rebuilding trained compressions of key-value caches and serving specifically-phrased updates beside a memory, as pasted text or injected cache state, show particular promise for keeping precomputed memories current, the latter as an interim measure between rebuilds, and we measure the cost and name the remaining questions associated with each.
Read the full abstract

Computation and Language

BiG-SURE - Bipartite Graph for Semantic Uncertainty and Reliability Estimation of LLMs

Reliable uncertainty estimation is a crucial requirement for deploying large language models (LLMs) and vision-language models (VLMs) in safety-critical settings, especially when the model parameters are not accessible (black-box). We propose BiG-SURE, an uncertainty estimator based on cross-temperature semantic agreement. The method samples low-temperature responses as stable semantic anchors and high-temperature responses as probes under meaning-preserving input transformations. It then constructs an anchor-probe Bipartite Graph (BiG) using NLI-based entailment scores and defines confidence through the normalized squared spectral energy of this matrix, with uncertainty given by its complement. This bipartite graph-based Semantic Uncertainty and Reliability Estimation (SURE) score measures whether high-temperature probes remain semantically aligned with the model's stable low-temperature belief or not. We evaluate BiG-SURE on text QA, multilingual QA, and multimodal QA tasks across multiple model families. In these experiments, BiG-SURE improves average abstention AUROC over prior black-box uncertainty estimators, while remaining simple, unsupervised, and applicable to black-box model settings.
Read the full abstract

Materials Science

Emergence and suppression of phonon vortices in two-dimensional crystals: Interplay of lattice symmetry, heavy impurities, and shear

Phonon vortices are vortex-like displacement fields that appear in the vibrational modes of two- dimensional materials. Here, we demonstrate that these vortices arise as symmetry-adapted linear combinations of degenerate planar phonon modes, with the superposition coefficients uniquely de- termined by the lattice point group. This symmetry principle establishes that vortices are intrinsic standing-wave solutions in perfect crystals, requiring neither impurities nor disorder. A heavy mass impurity favors vortex modes over planar modes through stronger resonance-induced frequency soft- ening, whereas shear deformation suppresses vortex modes by breaking rotational symmetry. The competition between these two effects gives rise to a precisely predictable strain threshold. The proposed framework, grounded in symmetry and energy-minimization, provides a useful basis for understanding vibrational topological defects induced by other types of impurities or defects. This study further suggests that defect and shear engineering constitutes an effective tuning strategy for controlling vibrational modes and the associated thermal and mechanical properties of crystals.
Read the full abstract

Statistical Methodology

Marginal Coordinate Test for Fréchet Regression with Random Objects

We develop a marginal coordinate test for regression with Euclidean predictors and a random-object response in a separable metric space. The goal is to test whether a predictor provides additional information about the response conditional on the remaining predictors. In a semi-supervised design, an unlabeled sample is used to estimate predictor conditional means, while an independent labeled sample is reserved for inference. The resulting residuals are combined with a product-space kernel to form a kernel conditional mean dependence (KCMD) U-statistic without requiring a response residual. The primary identity-based test targets a necessary conditional mean restriction, while a multiple-transformation extension probes broader alternatives. We establish a weighted centered chi-square null limit, wild bootstrap validity, consistency against fixed detectable alternatives, and local power under mean-element alternatives. For simultaneous inference, truncated p-to-e calibration combined with e-BH provides asymptotic false discovery rate control under general dependence. Simulations with Euclidean and non-Euclidean responses, together with a New York City taxi-flow analysis, illustrate the method.
Read the full abstract

All authors Jiaye Chen, Rui Qiu, Roulin Wang, Zhou Yu

Robotics

Temporal Forcing: 4D Representation Alignment for Vision-Language-Action Models

Long-horizon robotic manipulation requires vision-language-action (VLA) models to track scene states and their evolution beyond the current observation. However, simply conditioning policies on observation history does not guarantee that the history is effectively utilized: action supervision constrains what the policy should do, but only indirectly constrains what its history representations should retain. To resolve this, we present Temporal Forcing, a 4D representation alignment framework that explicitly supervises latent temporal states and their transitions. Specifically, we first introduce a history pathway that compresses past observations into compact latent tokens. We then align these tokens and current-frame features with geometric targets from a pretrained 4D foundation model, providing direct supervision at both the state and transition levels. The 4D foundation model and alignment heads are used only for training-time supervision. Temporal Forcing improves average success from 96.6% to 98.8% on LIBERO, with the largest gain on LIBERO-Long (93.8% to 97.2%), and from 53.5% to 62.8% across twelve RoboTwin 2.0 tasks. Furthermore, Temporal Forcing increases full-task success from 20.0% to 43.3% on a physical multi-stage hidden-placement task. Controlled experiments show that 4D representation alignment is crucial for making observation history beneficial to the model. Code will be publicly available.
Read the full abstract

All authors Xingyu Ding, Yuzhong Zhao, Chunhai Zhao, Yinghuan Shi, Chaoyang Zhao, Yifan Zhang

Materials Science

Topological and spin-orbit effects on orbital moments in ultra-thin magnetic films

Topological orbital moments (TOMs) are a direct hallmark of a magnetic texture with a non-trivial spin topology. In addition to giving insight into the topology of the magnetic texture, TOMs could also be used to manipulate magnetic structures with a compensated total spin moment. Experimental evidence of TOMs has been provided via transport measurements of the Hall effect in intercalated van-der-Waals materials. However, a direct observation of TOMs is still missing. Another complication arises for the unambiguous proof of the topological origin since orbital moments can also occur due to spin-orbit coupling. Here, we use first-principles electronic structure theory to investigate the origin of orbital moments in different compensated spin structures with a non-trivial topology. We focus on ultrathin magnetic films at surfaces such as Pd/Mn bilayers on Re(0001) which represent ideal model systems for the detection of TOMs since it is possible to apply experimental techniques with local resolution of magnetic properties such as spin-polarized scanning tunneling microscopy. Due to its trivial topology we use the row-wise antiferromagnetic (RW-AFM) state in a hexagonal monolayer to analyze the spin-orbit induced contributions. The triple-Q (3Q) state is a superposition state of three RW-AFM (1Q) states and thus electronically similar, however, due to its non-trivial spin topology it exhibits TOMs. The comparison between these two spin states allows us to disentangle the topological and spin-orbit contributions to the orbital moments. We find that TOMs have an important contribution in spin-compensated systems, since the spin-orbit coupling induced orbital moments are nearly compensated. First-principles calculations for atomic-scale skyrmion lattices in Fe monolayers on different surfaces exhibit the same general trend found for the 3Q state.
Read the full abstract

All authors Felix Nickel, Soumyajyoti Haldar, Mara Gutzeit, Stefan Heinze