Computation and Language
Agent-Reflex: Teaching Language Agents to Act Reflectively
Guanting Dong, Xinyu Yang, Jiajie Jin et al. (+9 more)
Large
language models have evolved from single-turn chatbots into interactive
agents capable of long-horizon
reasoning. In real deployments, MCP tool services can return noisy feedback and users may intervene to correct the
agent, but existing methods provide limited supervision for reflection learning and failure recovery. To bridge this gap, we propose
Agent-Reflex, a general training framework for incentivizing proactive reflection in
agents. Specifically,
Agent-Reflex comprises three stages. First, Environment Reflection Skill Discovery (RSD) injects grounded tool-feedback perturbations into environment scaling, samples a capability-diverse LLM pool, mines matched success-failure pairs, and summarizes reusable reflection meta-skills. Building on this skill library, we then apply Reflective Supervised Fine-Tuning (ReSFT) that converts each skill into supervision for self-reflection, skill recovery, and user-centric correction, strengthening complementary reflective behaviors. Finally, we propose On-Policy Reflective Policy Optimization (ORPO) to combine agentic reinforcement learning under noisy interactions with reflection-guided on-policy
distillation to further elicit proactive reflection. Built on Qwen3-8B,
Agent-Reflex reaches 58.6 overall on five long-horizon tool-use
benchmarks and 29.1 on AgentNoiseBench, surpassing strong environment-scaling and reflection baselines. Experiments on six challenging
agent benchmark suites demonstrate the effectiveness of
Agent-Reflex. Further analyses show that
Agent-Reflex improves
agents' robustness to noise while maintaining stable training dynamics, supporting scalable
agent training. Our code is available at https://github.com/RUC-NLPIR/
Agent-Reflex.
Read the full abstract ▾
All authors Guanting Dong, Xinyu Yang, Jiajie Jin, Yuyang Hu, Xiaoshuai Song, Yifei Chen, Xiaoxi Li, Jiejun Tan, Yutao Zhu, Tat-Seng Chua, Ji-Rong Wen, Zhicheng Dou
Cryptography and Security
Standing, Not Truth: Who May Say What About Whom in Agentic Selection
Yedidel Louck, Amit Dvir and Ariel Stulman
Software
agents increasingly choose on a user's behalf among competing offerings described by the parties who benefit. A lawful participant, publishing under its own identity, compromising no channel and issuing no instruction, can move that choice with a single false factual claim about a competing offering. Across twelve published defenses spanning the three placements a check can occupy, none separates it from marketing. Each accepts it for a reason correct on its own: its origin is a lawful channel, its influence is evidence meant to decide, its action serves the user's goal. None establishes that the publisher was entitled to speak about a competitor, each exclusion stated by the defenses' own authors. Disclosure does not replace enforcement: re-rendering every description as attributed
speech leaves the attack at 72/72, the
agent repeating the attribution and
reasoning from it. The missing relation is standing: whether a claimant's bound identity entitles it to assert about a claim's subject. A publisher may say what its own package does. It has no standing to say a competitor's package is frozen, and the difference is decidable without settling whether either is true. Deciding it needs the one fact prose extraction does not recover, who spoke, which a catalogue binds to every entry and then drops. The check is quiet on honest traffic, flagging 0.9% of natural pairs in two corpora. Used as a gate it stops 130 of 131 landed attacks and disturbs none of 348 honest selections. Applied everywhere it changes 0.9% and 2.6%, where deleting the whole description changes 12.7% and 21.0%. On thirteen further
agents the attack succeeds on 50.7% of decisions and sentence removal leaves 10.3% standing. Two published attacks stay outside: self-directed claims, which standing admits, and injected criteria, with no subject in the set.
Read the full abstract ▾
Computer Vision
FETALFusion: Resolution-Aware State-Space Modeling for Multi-Domain Fetal Head Segmentation
Ammar Bhilwarawala and Mainak Bandyopadhyay
Head circumference (HC) measured from 2D ultrasound is a routine marker of fetal growth, and automating it needs segmentations whose boundaries hold up on individual scans, not only on average. We present FETALFusion, a dual-path CNN-Mamba encoder-decoder for fetal head segmentation across acquisition types. It applies four-directional state-space scans only at the two highest-resolution stages, routes bottleneck features through domain-specific channel gates that are averaged at test time, and adds a weak landmark shape prior. We train jointly on transabdominal (HC-18) and transperineal (PSFHS) images and compare ten methods on a held-out test set of 471 images, each trained under one schedule and scored by surface distances at native resolution. FETALFusion reaches a Dice score of 0.963 and an HD95 of 2.10 mm without test-time augmentation or a domain label. It improves on U-Mamba on all five reported metrics (HD95 reduced by 0.31 mm; Holm-corrected ($p < 10^{-4}$), a lead that holds under test-time augmentation and across two independent training runs, and in the mean on U-Net,
Attention U-Net and TransUNet. Residual encoders close the gap: a ResNet-34 UNet++ exceeds FETALFusion, and an ImageNet initialised ResNet-34 U-Net is best overall (HD95 1.75 mm). The ablation agrees. Calibrated against run-to-run variation, it attributes FETALFusion's measurable gains to the landmark shape prior and, on transperineal images, to the domain gates, while removing the state-space branch does not measurably change accuracy. HC from the predicted masks agrees with the clinical reference at an intraclass correlation of 0.9986.
Read the full abstract ▾
Optics
Recovery-Oriented Optical System Design Based on Fisher Information
Zhuang Qian, Yan Mo, Hao Tan et al. (+5 more)
Traditional optical system design for imaging applications has long been guided by image quality metrics such as geometrical RMS spot size, wavefront error, and modulation transfer function (MTF). Although these metrics effectively characterize direct imaging performance, they do not necessarily indicate optimal behavior when image restoration is subsequently applied. In modern imaging pipelines, however, optical systems are increasingly combined with physics-based or learning-based restoration methods, making the information transfer properties of the optical system critically important. In this work, we propose a recovery-oriented optical design approach that explicitly links optical system optimization with image restoration performance through an estimation-theoretic framework based on Fisher information. A new design criterion is adopted to ensure that the information encoded by the optical system can be reliably recovered during image restoration. To validate the proposed approach, two dual separated aspheric lenses are designed under identical configurations using conventional RMS-based optimization and the proposed recovery-oriented optimization, respectively. Although the RMS-optimized lens exhibits superior direct imaging quality, it underperforms after image restoration compared to the recovery-oriented design. These results demonstrate that recovery-oriented optical design provides a physically interpretable and effective extension of traditional optical design criteria for imaging systems operating within modern computational imaging pipelines.
Read the full abstract ▾
All authors Zhuang Qian, Yan Mo, Hao Tan, Yu Chen, Haibo Wang, Weining Guo, Yijie Ni, Donglin Ma
Information Theory
Latency-Aware Uplink NOMA-MEC with PASS: A Feasibility-First Resource Allocation Framework
Sepehr Abolhasani, Saba Asaad, S. Mohammad Razavizadeh et al. (+1 more)
Meeting hard task-completion deadlines for many simultaneously offloading users is the binding challenge in mobile-edge computing (MEC). As the user count grows, the transmit power required to satisfy every deadline within per-user budgets rises until no feasible allocation exists. We take a feasibility-first view of this $K$-user problem in an uplink non-orthogonal multiple access (NOMA) MEC system served by a pinching-
antenna system (PASS), whose radiating points establish strong line-of-sight (LoS) links that lower the received-power requirement. We formulate a latency-constrained joint optimization of transmit powers, CPU frequencies, and
antenna configuration. We then derive a closed-form least-power feasibility test: the componentwise-least allocation meeting every deadline. It certifies feasibility with the largest attainable margin to the power budgets and defines a practical quality-of-service (QoS)-feasibility outage metric, the probability that no allocation meets all deadlines within budget. We solve the problem with a three-block alternating-optimization
algorithm that is provably monotone-convergent and has
polynomial complexity. Numerical results show that the proposed scheme sustains deadline feasibility at user loads where equal-power, orthogonal multiple access (OMA), and single-
antenna baselines fail or degrade sharply, each by a mechanism predicted directly from the system model, whereas the proposed design degrades gracefully. Stress tests further show that feasibility is sustained under order-of-magnitude budget reductions, and that the pinching-
antenna array itself enables constrained-budget operation. The result is a scalable, closed-form-initialized design for deadline-feasible offloading.
Read the full abstract ▾
All authors Sepehr Abolhasani, Saba Asaad, S. Mohammad Razavizadeh, Jiguang He
Machine Learning
HaloBlock: Faster and Better Decoding for Diffusion Large Language Models
Zhaohui Zheng, Chenhang He, Jinrui Zhang et al. (+2 more)
Diffusion large
language models (dLLMs) have attracted increasing research
attention due to their bidirectional
attention and arbitrary order decoding. However, there are two issues in dLLMs inference: the decoding difficulty is dominated by the
tokens on the right side of the block, and the block-wise decoding inherently suffers from boundary contextual discontinuity. To tackle these limitations, we propose HaloBlock, a difficulty-aware adaptive
halo mechanism for dLLM inference. We first leverage bidirectional cumulative
entropy to characterize block decoding difficulty. Then, based on the bidirectional decoding difficulty, we adaptively allocate bidirectional
halo regions on both sides of each block to mitigate boundary contextual discontinuity. The difficulty-aware design of HaloBlock smooths cross-block context propagation and lowers decoding difficulty for hard
tokens, yielding better generation quality and faster convergence. Evaluated on LLaDA-8B-Instruct, Dream-7B-Instruct and LLaDA-1.5 across mathematical
reasoning and code generation
benchmarks, HaloBlock demonstrates consistent accuracy improvements and higher inference efficiency. Source codes are publicly available at https://github.com/PolyU-VCLab/HaloBlock.
Read the full abstract ▾
All authors Zhaohui Zheng, Chenhang He, Jinrui Zhang, Ming-Ming Cheng, Lei Zhang
Instrumentation and Detectors
Performance and Radiation Damage of the Heavy-Ion Silicon Beam Tracker in the BM@N Experiment
D. Chemezov, S. Khabarov, Yu. Kopylov et al. (+6 more)
The Silicon Beam Tracker (SiBT) of the BM@N experiment at the NICA complex (JINR, Dubna) is designed to reconstruct trajectories of beam ions upstream of the target. It consists of three stations based on double-sided silicon strip detectors (DSSDs) installed directly in the relativistic heavy-ion beam. The SiBT performance and radiation damage were studied using data from the 2026 experimental run with $^{124}$Xe beams at~$1.6$,~$2.2$, and~$3.0$~AGeV. During the run, the leakage current of the DSSD of the third station of SiBT increased by $28.95~\mu\mathrm{A}$, while the charge collection efficiency in the beam center decreased to approximately $68\%$ at the nominal bias voltage of 30~V. Increasing the bias voltage to 90~V restored the charge collection efficiency to approximately $95\%$. Despite the radiation-induced signal degradation, the tracking efficiency remained close to $99\%$. The accumulated radiation damage was also used to reconstruct the spatial distribution of the integrated beam fluence from strip-by-strip leakage-current measurements, yielding an average fluence of approximately $1.3\times10^{12}~\mathrm{n_{eq}/cm^2}$.
Read the full abstract ▾
All authors D. Chemezov, S. Khabarov, Yu. Kopylov, I. Osokin, A. Sheremeteva, E. Streletskaya, O. Tarasov, E. Zubarev, N. Zamiatin
Statistics Theory
Optimal scaling and curvature correction for constrained MCMC on spheres
Nayef Shkeir and Tobias Grafke
Sampling Gaussian measures restricted to nonlinear
manifolds in high dimensions requires a proposal kernel that respects both
geometric constraints and dimensional scaling requirements. We introduce tangent preconditioned Crank-Nicolson (pCN), which combines a
manifold projection with a cancellation of the ambient Gaussian contribution, leaving only a
geometric residual. Considering a shifted sphere with $\sqrt d$ scaling of radius and centre as a tractable
benchmark embedded
manifold, we prove that uncorrected tangent pCN results in a proposal noise scale of order $d^{-1/6}$ (compared to $d^{-1/2}$ for a tangent random-walk). We establish asymptotic optimal tuning for the geodesic expected squared jump distance which is maximised at an acceptance probability of 0.574 for the family of drift-corrected tangent proposals containing tangent pCN and tangent MALA. We also prove a joint
diffusion limit for the parallel and orthogonal coordinates to the sphere's centre vector of tangent pCN. The orthogonal
diffusion speed matches the limiting geodesic expected squared jump distance and is maximised at acceptance probability 0.574, whereas the parallel speed is maximised at 0.712, demonstrating that optimal tuning depends on the observable. We then identify an intrinsic precision correction characterised by potential Hessian and Ricci
curvature that yields a proposal kernel with dimension-independent fixed step size limit. We discuss a generalisation to arbitrary hypersurfaces and present nondegenerate scaling of tangent pCN applied to conditioned
diffusions under resolution refinement.
Read the full abstract ▾
Analysis of PDEs
Existence of solution for Schrödinger-Newton system with vanishing magnetic potential
Liying Shan, Wei Shuai, Ping Yang et al. (+1 more)
In this paper, we study the following Schrödinger-Newton system with a vanishing
magnetic potential and an electric potential \begin{equation}\label{0.1} \left\{ \renewcommand{\arraystretch}{1.25} \begin{array}{ll} (-i\nabla +A(x))^2u+V(x)u=\Phi u, \ & x\in\mathbb{R}^3,\\ -\Delta \Phi =|u|^2, \ & x\in\mathbb{R}^3, \end{array} \right. \end{equation} where $i$ is the imaginary unit, $A:\mathbb{R}^3\to \mathbb{R}^3$ is a continuous
magnetic potential, and $V:\mathbb{R}^3\to \mathbb{R}$ is a continuous electric potential. First, we show that the ground state energy of system with $A(x)\equiv0$ and $V(x)\equiv V_{\infty}>0$ is isolated. Using this fact together with a global compactness lemma, a barycenter function, and Brouwer degree theory, we prove that under suitable conditions on $A$ and $V$, system admits either a ground state solution or a high energy solution. Interestingly, whether a ground state solution exists is significantly affected by the
magnetic potential $A$ and the electric potential $V$. Furthermore, we provide sufficient conditions that guarantee the existence of a ground state solution, as well as conditions under which no ground state solution exists and a high energy solution appears instead.
Read the full abstract ▾
All authors Liying Shan, Wei Shuai, Ping Yang, Jianghua Ye
Computer Vision
AI-Enhanced Hyperspectral Imaging for Intraoperative Detection of Low-Grade Glioma: A Clinical Study
Siri Willems, Ishika Jain, Jens De Winne et al. (+9 more)
Current image-guided neuronavigation techniques, such as intraoperative MRI and ultrasound, are limited by high costs, limited sensitivity and operational complexity. There is a clear need for more efficient systems to enhance intra-operative tumor detection and improve patient outcomes for intrinsic
brain tumor resection. Whereas previous intraoperative
brain HSI studies have predominantly investigated high-grade gliomas, glioblastomas, or mixed-grade cohorts, this clinical study focuses specifically on low-grade gliomas (LGGs). LGGs strongly resemble normal
brain under white light conditions on the operative microscope. This study characterizes spectral differences between healthy and tumor tissue and provides a paired comparison of HSI- and RGB-based classification at matched tissue locations and under identical patient-level data splits. Tumor regions showed consistently lower reflectance, which may be associated with differences in hemoglobin concentration and oxygenation. A range of classification models was evaluated. The 3D CNN, which integrates spectral and spatial features, achieved the highest median ROC-AUC of 74.5%, compared with 67.9% for the spectral-only XGBoost model and 64.0% for the RGB-based model. These results support further investigation of HSI for low-grade glioma classification and quantify its performance relative to conventional RGB imaging in this cohort.
Read the full abstract ▾
All authors Siri Willems, Ishika Jain, Jens De Winne, Siri Luthman, Hamed Zivari Adab, Roeland Vandebriel, Anna Alperovich, Xiaohan Zhang, Ben Stoffelen, Tom De Schepper, Lien Solie, Steven De Vleeschouwer
Mesoscale and Nanoscale Physics
WannierSTM: Efficient superconducting scanning-probe spectroscopy from maximally localized Wannier functions
Oscar Moreno Segura, Dorye L. Esteras, Andrei Shumilin et al. (+3 more)
We present a Python-based program that combines maximally localized Wannier functions-based first-principles electronic structure with the nonequilibrium Green's function formalism to enable efficient simulations of scanning tunneling microscopy (STM). The workflow builds upon standard density functional theory packages to construct compact transport
Hamiltonians that preserve the properties of realistic materials. Superconductivity is incorporated via an effective pairing potential within the Nambu-spinor formalism, while coupling to the STM tip and substrate electrodes is treated using the wide-band limit approximation. Within this formulation, transport quantities such as current-voltage characteristics and differential conductance are evaluated in a closed analytical form. This avoids explicit numerical energy integrations and enables rapid spectroscopic calculations over broad parameter spaces. The framework provides a unified and computationally efficient approach for modeling superconducting hybrid nanostructures with
spin-orbit coupling,
magnetic textures, and
spin polarization. Representative applications in
spin chains, two-dimensional and molecular magnets on a
superconductor demonstrate its robust capabilities for both simple model systems and realistic ab initio molecular structures relevant to modern STM experiments.
Read the full abstract ▾
All authors Oscar Moreno Segura, Dorye L. Esteras, Andrei Shumilin, Gonzalo Rivero-Carracedo, Riku Tuovinen, José J. Baldoví
Numerical Analysis
Space-Time Parallel Direct Numerical Simulation of Rayleigh-Bénard Convection
Thomas Saupe, Thibaut Lunet, Mathis Bode et al. (+3 more)
Rayleigh-Bénard convection (RBC) is a model for turbulent flow and challenging to simulate by direct numerical simulation (DNS). Adequate resolution of all relevant length scales requires very fine grids in space and stability restrictions can dictate small step sizes. In this paper, we investigate the use of parallel spectral deferred correction (SDC) as a high-order time-stepping method to reduce wall-clock times of direct numerical simulation of RBC. Performance results on JUSUF (CPUs) and JUPITER (GPUs) with two different codes show parallel SDC can offer some concurrency along the temporal axis plus larger maximum stable time steps. Taken together, this allows for a reduction of wall-clock time required for simulations over one free-fall time by having to compute fewer steps and being able to mitigate the higher cost per time step through temporal parallelization.
Read the full abstract ▾
All authors Thomas Saupe, Thibaut Lunet, Mathis Bode, Robert Speck, Sebastian Götschel, Daniel Ruprecht
Analysis of PDEs
Cubic Nonlinear Schrödinger Systems and Orthonormal Fermionic Ground States in $\mathbb R$
Bin Chen, Yujin Guo, Yong Luo et al. (+1 more)
We study the one-dimensional nonlinear Schrödinger system \[ -u_i''-2\Big(\sum_{k=1}^N u_k^2\Big)^{p-1}u_i=\mu_i u_i \quad \mbox{in } \ \mathbb R,\ \ i=1,2,\cdots,N, \tag{S} \] where $u=(u_1,\cdots,u_N)\in H^1(\mathbb{R})^N$, $\mu_1\leq\mu_2\leq\cdots\leq\mu_N<0$, $1<p<3$ and $N\geq 2$ are arbitrary. For the cubic case where $p=2$ and $N\ge 2$, we establish the following structural results of the system $(S)$: (i) all nontrivial solutions can be classified completely; (ii) the linearized operator at any nontrivial solution is non-degenerate; (iii) exact $L^2$-mass formulas of all nontrivial solutions are derived in terms of distinct Lagrange multipliers $\mu_i$. As the applications of the above structural results, we first show that the system $(S)$ with $p=2$ does not have any orthonormal solution for any $N\ge2$. We then investigate a fermionic variational problem $J_N(p)$ with $N\ge2$, whose minimizers correspond to ground states of the system $(S)$ under orthonormal constraints. More precisely, we prove that there exists a sufficiently small $\delta_N>0$ such that $J_N(p)$ is attained for any $p\in(2-\delta_N,2)$, whereas $J_N(p)$ is not attained for any $p\in[2,2+\delta_N)$. These settle some
conjectures of [Gontier–Lewin–Nazar, ARMA, 2021] and [Frank–Gontier–Lewin, CMP, 2021].
Read the full abstract ▾
All authors Bin Chen, Yujin Guo, Yong Luo, Juncheng Wei
Numerical Analysis
Block diagonal preconditioners for arrow-shaped multiple saddle-point linear systems
Luca Bergamaschi and Michele Bergamaschi
We consider multiple saddle-point linear systems with block-arrow structure. We analyze exact and inexact Schur complement based block diagonal preconditioner, by characterizing the
eigenvalues of the preconditioned
matrices in terms of the zeros of specific rational functions. Numerical results onto large tests coming from
liquid crystal modeling and PDE-constrained optimization validate the obtained
eigenvalue bounds
Read the full abstract ▾
Machine Learning
Sea-Net: A Small and Inherently Interpretable Neural Network for Time Series Classification
Ubaid Ur Rehman and Tanmoy Mondal
Traditional time-series classification methods often operate as black boxes, providing a class prediction without indicating which time point or time interval within the series led to that prediction. To improve temporal interpretability, MILLET (Multiple Instance Learning with Temporal Explanation), which combines an InceptionTime encoder with Conjunctive Pooling, provides both classification and temporal explanations. However, MILLET contains approximately 424 K parameters, making it unsuitable for deployment on resource-constrained micro-controllers, and it achieves an overall time-series accuracy of 0.888 and an explanation score of 0.681 NDCG@n. We therefore propose SEA-Net, a lightweight architecture using small multi-scale depthwise-separable convolutional TSC encoders and improved Multiple Instance Learning (MIL) pooling heads. Our SEA-Net bottleneck encoder with Top-k pooling achieves 0.931 accuracy and 0.765 NDCG@n with only 41 K parameters. Another variant, using a gated encoder with Top-k pooling, achieves 0.955 accuracy and 0.750 NDCG@n with 62 K parameters. These results demonstrate that lightweight architectures can maintain strong classification performance and temporal explanations while substantially reducing model size. The proposed variants address specific limitations of the pooling strategy used by MILLET, providing a practical direction for interpretable time-series classification on resource-constrained edge devices.
Read the full abstract ▾
Differential Geometry
A stationary harmonic map with a nonstationary tangent map
Siran Li and Zijun Wang
We construct a stationary weakly harmonic map from a domain in \R^5 into a compact Kähler surface for which a tangent map is weakly harmonic but nonstationary. The target
manifold is the blow-up of \CP^1\times\CP^1 at four points lying in one fiber and arranged at the vertices of a regular tetrahedron. The nonstationarity of the tangent map arises from the break down of the tetrahedral
symmetry in the defect measure concentrated on the vertices.
Read the full abstract ▾
Computer Vision
Haoyang Jiang, Zhuotong Fan, Mingyang Yi et al. (+3 more)
Table
reasoning with
language models typically assumes access to clean textual tables. In practice, tables are often serialized into text, which can corrupt
cell contents, distort structural relationships, and discard visual cues, leaving the model to reason over incomplete evidence. In contrast, table images retain this information, but using it requires both accurate perception and
reasoning grounded in the relevant
cells. To address this, we propose MERIT (
Multimodal Evidence
Reasoning over Image Tables), a three-stage framework for learning to reason over table images, with serialized tables serving as optional auxiliary inputs (if provided). Our analysis identifies visual evidence acquisition as a bottleneck, motivating table-specific encoder adaptation through reconstruction, localization, and color-conditioned supervision. Building on the adapted representations, supervised fine-tuning teaches explicit evidence selection and
reasoning using verified trajectories. Finally, reinforcement learning refines the policy with separate rewards for evidence correctness, calculation, final answers, and output format. By providing feedback on intermediate steps, the framework targets
retrieval and computational errors that final-answer supervision alone does not distinguish. We evaluate the approach on table question answering and fact verification, analyze the contributions of visual inputs and training components, and assess generalization to additional
datasets with different table structures and visual presentations.
Read the full abstract ▾
All authors Haoyang Jiang, Zhuotong Fan, Mingyang Yi, Ju Fan, Lanqing Hu, Xiao Zan
Computer Vision
PUGS: Physics-Guided Uncertainty Modeling for Underwater Dynamic 3D Gaussian Splatting
Tingjun Li, Yuxi Wang, Guoqiang Zhong et al. (+1 more)
Dynamic 3D Gaussian Splatting (3DGS) provides an efficient framework for dynamic scene reconstruction. Most existing approaches model scene dynamics with deterministic deformation fields, assuming each Gaussian follows a distinct
geometric trajectory over time. This assumption breaks down underwater, where light scattering, refraction, marine snow and low visibility create substantial physical ambiguity in captured images. We present PUGS, a physics-guided uncertainty model for underwater dynamic 3DGS. PUGS learns inherent uncertainty induced by underwater image formation and leverages it as a physical signal for dynamic scene modeling. Rather than estimating a single deterministic displacement, it parameterizes each Gaussian's motion as a probability distribution with expected deformation and physical uncertainty. We adopt
stochastic deformation optimization to sample Gaussian trajectories via local reparameterization, exploring multiple physically plausible solutions during training. We further propose uncertainty
stochastic annealing and a dual-path self-consistency
distillation objective, converting
stochastic exploration to deterministic reconstruction while preserving temporal coherence. Extensive experiments on underwater dynamic
datasets and physics-aware
benchmarks demonstrate superior rendering quality, temporal consistency and reconstruction robustness. Our method achieves 1.16 dB PSNR gain over the strongest baseline on UVEB and 3.99 dB on DRUVA.
Read the full abstract ▾
All authors Tingjun Li, Yuxi Wang, Guoqiang Zhong, Zhaoxiang Zhang
Cryptography and Security
Provenance and Authorization in Autobiographical Language Models
Faruk Alpay and Taylan Alpay
Autobiographical
language models may draw on archives containing observations, quotations, emotional reports, and earlier model outputs. Cryptographic authentication can preserve an archive while leaving these sources semantically conflated. We formulate this gap as a computer-security problem: narrative content must not silently become consent or alter an authorized decision rule. We define typed records, authenticated attestations, a separate policy channel, and admissible views. Under explicit assumptions, we give a standard integrity reduction and a noninterference argument, characterize the information required by exact decision-preserving summaries, and show why passive history and fixed-point convergence need not identify a preferred action. Alpay
Algebra supplies a comparative
language for iterative
memory without being a premise of the proofs. Sociology of religious affiliation, source-monitoring psychology, and self-determination theory motivate distinctions in a bounded Latter-day Saint vignette. A separate, reproducible pilot uses a frozen public
language model on 32 wholly synthetic cases across 576 paired conditions. It compares flattened records, source labels, and an explicit admission gate under contradictory summary repetition. The study concerns computational attribution and authorization, not the motives or psychology of the people in the vignette.
Read the full abstract ▾
Information Theory
Chordal Codes: Code Construction and Efficient Gaussian-Elimination Decoding
Yulin Chen, Laigang Guo, Tong Li et al. (+1 more)
Maximum-likelihood decoding of linear codes over erasure channels can be considered as solving a linear system of a column submatrix of the parity-check
matrix. In practice, Gaussian elimination is usually applied, and its computational cost depends on the structure of the
matrix, especially the sparsity and the fill-in generated during the elimination. The chordal graph, or triangulated graph, first introduced by C. Berge, has a so-called perfect elimination ordering, indicating that if the primal graph of a
matrix is a chordal graph, then there exists a perfect elimination ordering such that a special class of fill-in under Gaussian elimination following this ordering will be avoided. In this paper, we introduce chordal codes, i.e., codes equipped with parity-check
matrices such that the primal graphs of the parity-check
matrices are chordal. We prove that, for a chordal code, the decoding submatrix associated with any erasure pattern retains a chordal primal graph. So, the complexity of chordal elimination (Gaussian elimination following the perfect elimination ordering applied to the chordal code decoding) will be smaller than decoding a random code by Gaussian elimination. Also, we provide an upper bound on the complexity of chordal elimination.
Read the full abstract ▾
All authors Yulin Chen, Laigang Guo, Tong Li, Mingyang Zhu
Computer Vision
HierarchiRetina: A Diabetic Retinopathy Screening and Grading Cascade with Heterogeneous Mixture-of-Experts Lesion Segmentation
Md Asif Uddin, Niloy Farhan, Rafeed Rahman et al. (+1 more)
Clinical screening for diabetic retinopathy proceeds in stages, yet most automated systems grade fundus images with a single black-box network. Three gaps remain: (i) lesions differ widely in size and contrast, but segmenters typically apply the same blocks to all; (ii) ungradable images are usually excluded or treated as an extra class, and lesion-aware graders seldom test whether lesions drive the grade; (iii) staged systems often report each stage only on the images reaching it, leaving propagated errors unmeasured. HierarchiRetina, a screening, segmentation and grading cascade with inspectable stage outputs, addresses each gap. First, HSMoE-AUNet is a U-Net whose skip, bottleneck and decoder blocks are sparse mixtures of heterogeneous experts, with lesion-matched types, counts and losses; SwinHRUNetPP, a dual-encoder vessel segmenter, uses a field-of-view-restricted loss. Second, LG-DRG separates gradability from ordinal severity and lets lesion masks query the image through gated cross-
attention. Third, an end-to-end evaluation counts screening misses as errors and traces each error to its stage. Pooled Dice ranged from 0.184 (microaneurysms) to 0.579 (exudates); vessel Dice was 0.731. On 58,689 held-out images from six
datasets, end-to-end quadratic weighted kappa was 0.804, with referral sensitivity 0.832 at specificity 0.947, and the screening gate caused 66% of errors. In ablations, the grade ignored the masks, even with the cross-
attention gate started open. On all 4,105 official DDR test images, six-class accuracy was 85.6%, above the best published six-class result we found (83.8%). Staged screening systems should be evaluated end-to-end, with stage attribution and mask ablation, before clinical deployment.
Read the full abstract ▾
All authors Md Asif Uddin, Niloy Farhan, Rafeed Rahman, Md. Golam Rabiul Alam
High Energy Physics (Experiment)
Real-time monitoring of LHCb beam spot properties based on FPGA hit reconstruction
Giulio Cordova, Elena Graverini, Daniele Passaro et al. (+3 more)
We present a real-time trackless method to monitor the centre of the LHCb luminous region and the relative
alignment of the two halves of the LHCb Vertex Locator, using cluster counts accumulated in selected regions of the detector and processed in firmware. A Principal Component Analysis performed on simulated data yields a set of weight vectors that reduce the position
estimators to scalar products of the counter rates, executable with negligible computational overhead. With pp collision data from 2024, the transverse position of the luminous region is reconstructed with a statistical precision of approximately 4 micrometre per coordinate in a 90 millisecond accumulation window. A measurement of the longitudinal centre is demonstrated with a statistical precision of a few millimetres in simulated studies. The introduction of run-time configurable firmware counters has been studied in simulation to improve the transverse and longitudinal resolutions by a factor of three with respect to the fixed-geometry baseline.
Read the full abstract ▾
All authors Giulio Cordova, Elena Graverini, Daniele Passaro, Federico Lazzari, Michael J. Morello, Giovanni Punzi
Earth and Planetary Astrophysics
Radiation limits on amino acid preservation and chiral measurement at Oxia Planum
Dimitra Atri
Amino acid enantiomeric excess (EE) is a potential biosignature, but its measurement requires sufficient surviving material as well as preserved chirality. We derive a criterion that separates abundance-limited from racemization-limited preservation and apply it to the Mars Organic Molecule Analyzer (MOMA) on the ExoMars Rosalind Franklin rover. The model combines
cosmic-ray and radiogenic doses with laboratory destruction kinetics and assumed ratios of interconversion to destruction rates. For an initially homochiral population of 1000~ppmw, the surviving EE is 58% or 25% when the total abundance reaches 1~ppmw for rate ratios of 0.04 or 0.10, respectively. This abundance criterion does not ensure that both enantiomers can be quantified. Applying provisional estimates of 1~ppmw per enantiomer and 10% EE, serine at 2~m requires initial concentrations of 340 and 960~ppmw after
cosmic-ray exposures of $10^{7}$ and $10^{8}$~yr, respectively, under transferred fused-silica–perchlorate kinetics and a 1.57-MGy radiogenic dose. These concentrations exceed measured meteoritic serine abundances, but do not predict the initial Oxia inventory. For organics retained in a closed host since deposition 3.9~Ga ago, the modeled radiogenic dose is 1.6–5.6~MGy across the adopted compositions. In the lower-dose case, drilling to 2~m reduces the total dose by only a factor of 1.5 after $10^{7}$~yr. The boundaries depend on mineral kinetics and organic residence time; they distinguish preserved chirality from a measurable excess.
Read the full abstract ▾
Robotics
Locomotion with Online Awareness of Dynamics for Precision Footholds
Kunzhao Ren, Zhishuo Zhu and Xiaobin Xiong
Unknown payloads alter
robot dynamics and make precise contact execution difficult when foothold margins are small. We present LOAD, a perceptive locomotion framework that combines explicit inertial estimates with implicit dynamics features for narrow-beam traversal. Separate recurrent modules infer torso inertial changes and task-relevant history from proprioception, while a height-map teacher is distilled into a depth-based student. We extensively evaluate LOAD and its variants in simulation and on hardware across beam widths and payload configurations. The results show that implicit dynamics features accelerate training, while explicit inertial estimation further improves traversal success; combining both enables LOAD to approach the performance of a privileged policy supplied with ground-truth inertial parameters. Hardware experiments demonstrate traversal with unilateral payloads of 6 kg on a 10 cm beam and 9 kg on a 15 cm beam, and an evenly distributed 11 kg payload on a 15 cm beam. These results support combining physical estimates and history features for precision locomotion under unknown loads.
Read the full abstract ▾
Systems and Control
Enforcing a Strict Safety Margin in LLM Text Generation via Barrier Lyapunov Control
Anand Singh, Shubham Singh, Anjali Tiwari et al. (+1 more)
Large
language models are increasingly deployed without formal guarantees against toxic text generation. Existing control barrier function (CBF) based safety filters can keep the generation trajectory within a toxicity-free region, but do not enforce a positive distance from its boundary, allowing formally safe trajectories to approach the toxicity threshold arbitrarily closely and leaving little buffer against modelling errors or disturbances. We address this limitation with a Barrier Lyapunov Function (BLF) safety filter that enforces a prescribed clearance margin throughout generation while retaining the continuous-time
transformer dynamics, including feedforward layers. The resulting safe input set is a half-space defined by a constraint affine in the control input, and is non-empty under certified surjectivity and classifier-nondegeneracy conditions. For margin-compatible initial conditions we prove that the clearance margin is never violated and that the prescribed clearance is restored exponentially, yielding an explicit clearance floor, and show that, as the margin vanishes, the admissible trajectories satisfy the exponential second-order CBF condition. We further derive an explicitly computable sufficient disturbance budget under which the clearance guarantee degrades gracefully. Simulations on a small
transformer illustrate the resulting improvement in safety clearance.
Read the full abstract ▾
All authors Anand Singh, Shubham Singh, Anjali Tiwari, Katsuki Fujisawa
Artificial Intelligence
Learning How to Remember: Memory Modality Selection for GUI Agents
Taeyun Roh, Suhyeong Park, Gijin Kim et al. (+2 more)
Recent advances in
vision-
language models (VLMs) have substantially improved the capabilities of GUI
agents, while
memory augmentation has emerged as a promising approach for reusing knowledge from past interactions. However, existing
memory augmented GUI
agents typically adopt a fixed representation strategy, providing retrieved experience as a static modality. It remains unclear whether a single
memory modality is consistently optimal, or how modality preference changes across interaction steps. In this work, we conduct a systematic study of
memory modality in VLM based GUI
agents. Across GUI
benchmarks, we find that the preferred
memory representation is strongly step dependent. Motivated by the finding, we introduce SwitchMem, a routing framework that dynamically selects between textual and visual
memory representations at each step. Across three GUI
benchmarks, SwitchMem outperforms
memory augmented baselines that rely on static
memory representations. Our findings validate our claim that
memory modality is a step dependent decision rather than a fixed design choice. Code and project resources are available at https://txxnrd.github.io/switchmem
Read the full abstract ▾
All authors Taeyun Roh, Suhyeong Park, Gijin Kim, Wonjune Jang, Jaewoo Kang
Robotics
GLST: Global-Local Spatio-Temporal Memory Decomposition Framework for Online Chained-Goal Navigation
Yousef Liang, Ryan Ardalan, Yarden Goraly et al. (+6 more)
Autonomous navigation in real-world environments often requires long-horizon operation under partial observability and evolving goals. Existing mapless navigation approaches predominantly assume fixed goals throughout an episode, limiting applicability to practical scenarios where goals change online during execution. We study online chained-goal navigation, where goals evolve through existing-goal redirection or new-goal emergence without explicit transition signals. Under this setting, we find navigation may require either continued exploration or strategic revisitation, creating conflicting
memory requirements. We propose Global-Local Spatio-Temporal Belief Learning (GLST), a
memory-decomposition framework that separates latent beliefs into global-local spatial and long-short temporal representations for long-horizon
reasoning. We further introduce goal-conditioned rewards to balance exploration and exploitation under evolving goals. Extensive experiments in simulation across diverse terrains demonstrate improved chained-goal navigation performance over baselines. Real-
robot experiments using the Spot quadruped further validate our
memory decomposition framework.
Read the full abstract ▾
All authors Yousef Liang, Ryan Ardalan, Yarden Goraly, Chenhao Zhu, Yue Yin, Jonas Frey, Claire Tomlin, Marco Pavone, Shankar Sastry
Computer Vision
CineMem: Attention-Aligned Fine-grained Memory Retrieval for Multi-Shot Long Video Generation
Xiaoxue Wu, Xinyu Gao, Fang Yu et al. (+2 more)
Autoregressive
video generation relies on
memory retrieval to model long-range dependencies. This capability becomes particularly critical for multi-shot generation, where models must reconcile substantial visual changes across shot transitions with persistent semantic continuity. Existing methods tend to perform
memory retrieval at the frame level and share the same historical context across all query
tokens, overlooking the heterogeneous information needs and leading to redundant
retrieval and inefficient use of historical
memory. To bridge this gap, we introduce CineMem, an
attention-aligned fine-grained
memory retrieval framework that combines online
memory organization with query-adaptive
retrieval for autoregressive multi-shot
video generation. Specifically, guided by intrinsic
memory-access patterns revealed through
attention analysis, we propose an online-clustered KV cache as a
memory organization strategy over all historical
tokens, enabling semantic-aware
memory grouping and more precise
retrieval. In addition, we introduce a query-level
retrieval mechanism that provides broader access to historical information under a fixed
retrieval budget, thereby improving long-term
memory utilization. Together, these designs enable the multi-shot
video generator to preserve long-range consistency across shot transitions, particularly in revisitation scenarios, while approximating full-
memory access under a constrained
memory budget. To facilitate evaluation, we further introduce a suite of task-specific, VLM-assisted metrics for faithful inter-shot consistency assessment. Extensive experiments demonstrate the effectiveness of our framework.
Read the full abstract ▾
All authors Xiaoxue Wu, Xinyu Gao, Fang Yu, Yu Qiao, Xinyuan Chen
Machine Learning
Stein Variational Distributional Reinforcement Learning: Score-Guided Particles for Multivariate Return Distributions
Manuel Haußmann and Melih Kandemir
Decisions that weigh several reward signals against each other require the joint distribution of the return vector. Categorical critics scale exponentially in the reward dimension, and
quantiles have no canonical multivariate analogue. Multivariate distributional critics therefore represent each return distribution by M equal-weight particles trained with a discrepancy loss. Beyond their atoms, the particles provide no sampler and cannot resolve events rarer than 1/M. We propose Stein variational distributional RL (SVDRL). Its critic outputs Stein particles, which Stein variational gradient descent transports along a learned score field of the distributional Bellman target. We learn this field by bootstrapped denoising score matching across noise levels and tie it across the Bellman recursion by a consistency loss, which we prove is a non-expansive temporal-difference scheme in score space. The same field certifies the particles. Its reverse-
diffusion sampler satisfies a supremal Wasserstein policy-evaluation bound that holds for bounded, possibly
multimodal returns. The Stein particles inherit this bound up to a gap that we measure without ground truth, and the sampler answers tail-risk queries below the particle resolution at any sample size. Empirically, SVDRL is more accurate on rare events than particle critics with the same number of particles, and on par on MuJoCo continuous control.
Read the full abstract ▾
Statistical Methodology
Free-support Wasserstein Resampling in Sequential Monte Carlo
Mengyu Li, Cheng Meng and Jun S. Liu
In modern sequential Monte Carlo (SMC) applications, such as
diffusion model samplers, propagating a particle is expensive, so only a few particles can be afforded. In this regime, each resampling step decides which regions of the state space survive, so a better resampling scheme is worth extra computation. However, classical resampling methods only replicate existing particles and suffer from sample impoverishment, especially when the particle size is small relative to the dimension. To create new particles, we relax the unbiasedness of resampling and control the mean squared error via a Wasserstein criterion. Specifically, we propose Wasserstein-Dirichlet resampling (WDR), which constructs an equal-weight measure that optimally approximates the weighted particle distribution by solving a free-support Wasserstein barycenter problem. We then randomize the resulting optimal coupling by Dirichlet mixing. Theoretically, we provide non-asymptotic mean squared error bounds that quantify the bias-variance trade-off and show that the rate depends on the intrinsic dimension. A truncated variant preserves high-weight particles while applying WDR to the remaining mass, targeting mild weight degeneracy. Experiments on state-space models and conditional image generation demonstrate that WDR consistently improves estimation accuracy, stability, and sample diversity. The improvements persist at matched wall-clock cost when the propagation cost dominates.
Read the full abstract ▾