Robotics
EIDA: Execution-Interface Dynamics Adaptation for Real-to-Sim-to-Real Robot Navigation
Yiwei Qian, Shanze Wang, Qingyuan Hu et al. (+2 more)
Simulation-to-
robot transfer can fail when velocity commands produce motion and feedback that differ from those modeled during policy training. We present execution-interface dynamics adaptation (EIDA), which fits these responses from target-platform execution data without reconstructing actuator dynamics. A model of body-frame pose increments updates simulator geometry, while a separate model predicts the velocity feedback observed by the policy; a short history of velocity feedback is included in the policy input. The fitted models are used within a lightweight GPU-parallel simulator. On the full Jackal and Go2 validation sets, the fitted models reduced position and yaw prediction errors relative to the simulator's predefined motion model. Across 100
benchmark navigation environments evaluated in a separate physics-based simulator, EIDA achieved the highest success rate and
navigation score among the compared learned policies, both with and without global guidance. Feedback ablations further supported the need to match policy-facing velocity estimates. On a physical Unitree Go2, EIDA reached the goal without collision in all 20 static-scene trials, compared with 4 of 20 for the baseline. These results show that execution-interface adaptation can improve
navigation transfer without detailed actuator simulation.
Read the full abstract ▾
All authors Yiwei Qian, Shanze Wang, Qingyuan Hu, Xinming Zhang, Wei Zhang
Computation and Language
AGO AI Quality Gate: Evidence-First Release Decisions for Retrieval-Augmented Generation
Giulio Zeloni, Enrico Lo Conte, Salvatore Rionero et al. (+3 more)
Enterprises adopting
retrieval-augmented generation (RAG) face a recurring operational decision: promote, revise, or block a system version. The evidence is incomplete and the metrics come from fallible LLM judges. We report on AGO AI Quality Gate (AGO), an evidence-first quality-gate framework deployed in industrial RAG assessment engagements. AGO integrates four key components: a four-state decision model that treats missing data and judge errors as explicit outcomes; layered scoring combining deterministic checks, local guardrails, and structured LLM evaluation; a stratified beta-binomial gate that quantifies regression risk probabilistically; and a mandatory meta-evaluation protocol to validate the LLM judge before it influences decisions. Since engagement data is proprietary, we evaluate the judge layer on RAGBench, a public
benchmark of 100k annotated RAG traces across 12
datasets. On identical stratified test samples (N=1200 per judge), a low-cost judge (gpt-4.1-nano) detects non-adherent answers barely above chance (AUROC 0.603 [0.570, 0.634]), despite producing flawless protocol output, while gpt-4o reaches 0.783 [0.756, 0.807] – yet its per-domain performance still ranges from 0.62 to 0.88. A fixed-seed gate study spanning regression, no change, and improvement quantifies unsafe promotion, false-alarm cost, and improvement throughput. Under regression, the decision-grade profile reduces unsafe promotion to 22.2%-35.1%, against 29.3%-41.8% for a naive gate. These results support the design choices that judge quality must be measured per engagement and that point estimates alone are not a release decision.
Read the full abstract ▾
All authors Giulio Zeloni, Enrico Lo Conte, Salvatore Rionero, Giuseppe Santoro, Alessandro Rastelli, Fabio Sorrentino
Mesoscale and Nanoscale Physics
Composite fermions in ideal Chern bands
Songyang Pu, Liangtao Peng and Shaffique Adam
The composite-
fermion framework maps the strongly correlated fractional
quantum Hall state onto a weakly interacting integer
quantum Hall state in a reduced effective
magnetic field. We ask how this framework is modified for fractional Chern
insulators, where the role of a uniform
magnetic field is replaced by the nonuniform Berry
curvature of a Chern band. Specifically, we construct composite-
fermion wave functions for Jain states and their quasiparticles and quasiholes in Aharonov-Casher bands, a widely used model for ideal Chern bands. We find that these states overlap closely with exact eigenstates over a broad range of field modulation. Using these wave functions, we demonstrate that the emergent composite-
fermion band structure goes beyond the conventional picture of effective Landau levels, correctly capturing the quasiparticle dispersion. Remarkably, we find that an isolated quasiparticle band carries a total Berry flux that is different from a Landau level. These results establish a composite-
fermion description for mobile fractional Chern
insulator anyons and provide a framework to understand their stability.
Read the full abstract ▾
Metric Geometry
Morley Simplices in Higher Dimensions: Regularity, Reflections, and Counterexamples
Quang Hung Tran
Trisecting the dihedral angles of an $n$-simplex defines its Morley simplex. We study the original simplices for which this simplex is regular. A criterion in terms of the Gram
matrix of the facet normals reduces the problem to a
matrix equation. The derivative of the Morley map at the regular simplex has two explicit
eigenvalues, both nonzero for $n\ge3$; thus the regular simplex is an isolated solution. We
conjecture that in dimensions four and five a regular Morley simplex forces two hyperplane reflections interchanging disjoint pairs of vertices. We prove that this conclusion fails in every dimension $6\le n\le200$ and in every dimension $n=\binom k2-1$ with $k\ge9$: in these dimensions there are simplices with regular Morley simplex and no hyperplane reflection
symmetry. The examples for $8\le n\le200$ have dihedral
symmetry of order $2(n+1)$, while the infinite family is based on the Johnson scheme. In dimension four we give exact constructions of two nonregular examples, defined by irreducible
polynomials of degrees $18$ and $8$ with Galois groups $S_{18}$ and $S_8$. Neither example is expressible by radicals. The computer assisted existence proofs use exact rational arithmetic and intervals with outward rounding.
Read the full abstract ▾
Computer Vision
AutoGUIWorld: Image Generators as Visual World Models for GUI Agent
Cheng Yang, Yifan Wu, Yutao Huang et al. (+18 more)
GUI
agents require high-quality interaction trajectories to learn how software environments respond to actions, maintain state, and support multi-step workflows. However, the diversity of available trajectories is constrained by the applications, interface states, and workflows accessible in the underlying environments. Expanding this coverage requires deploying increasingly diverse and complex software, with specialized applications imposing additional installation, configuration, and runtime costs. We introduce AutoGUIWorld, a data generation framework that combines the visual priors of image generators with the task knowledge of a planner to synthesize GUI interaction trajectories without deploying or running the corresponding software environments. AutoGUIWorld samples initial GUI scenes from structured specifications of operating-system context, visual appearance, and interface state, and generates tasks conditioned on those scenes. A planner then specifies atomic actions and their intended visual consequences, while an image generator iteratively edits the current screenshot to produce subsequent observations. Action grounding and transition-level quality filtering yield 79,266 spatially annotated step-level training samples across Ubuntu, Windows, macOS, and Chrome. Fine-tuning Qwen3.5-35B-A3B on AutoGUIWorld trajectories improves the mean task score on OSWorld from 33.0% to 40.8% and the task success rate on ScienceBoard from 14.0% to 32.2%. These results show that generated trajectories improve GUI-
agent performance on real desktop and scientific tasks.
Read the full abstract ▾
All authors Cheng Yang, Yifan Wu, Yutao Huang, Zhaohua Zhang, Beiduo Chen, Muxi Chen, Chenchen Zhao, Hexuan Deng, Haolin Yang, Geyuan Zhu, Sa Zhu, Jianhuan Zhuo, Qiuyong Xiao, Jianhao Ruan, Yiran Peng, Jiayi Zhang, Tian Ye, Xinlei Yu, Tianwen Jiang, Jihong Zhang, Yuyu Luo
High Energy Astrophysical Phenomena
25 Years of Mrk 421 with XMM-Newton: Unveiling Structured Jets and Energy-Dependent Escape
Tek P. Adhikari, Gopal Bhatta, Sangeetha Kizhakkekalam et al. (+4 more)
We present a long-term X-ray spectral and timing analysis of the TeV blazar Mrk~421 based on nearly a quarter century of archival XMM-Newton/pn observations obtained between 2000 and 2025. Our sample comprises 76 epochs obtained with the pn instrument in both IMAGING and TIMING modes, allowing us to trace the source's variability and spectral
evolution. Mrk~421 exhibits flux variations by approximately an order of magnitude, with prominent flaring periods interspersed with low-activity states. A clear harder-when-brighter trend is observed, accompanied by scatter in the hardness ratio (HR), suggesting flux alone does not uniquely determine the spectral state. Fractional variability in the hard band ($2-10$ keV) correlates tightly with that in the soft band ($0.3-2$ keV), with a steeper slope indicating pronounced variability at higher energies. The flux distribution is non-Gaussian and lognormal-like, dominated by moderate flux levels with a tail of bright
flares. Spectral analysis shows that power-law model is not sufficient; 59 epochs are best described by a broken power-law (BPL) and 17 by a log-parabola (LP). This fluctuation between BPL and LP models suggests the emission region transitions between a distinct acceleration shockfront and downstream
stochastic turbulence. The spectral break energy of BPL remains predominantly clustered around $2$ keV across all flux states, while the average spectral steepening above the break is mild ($\Delta \Gamma \approx 0.19$), inconsistent with standard one-zone cooling models. Our results support structured jet scenarios or energy-dependent electron escape, providing key constraints on the long-term behavior of relativistic jets in high-synchrotron-peaked blazars.
Read the full abstract ▾
All authors Tek P. Adhikari, Gopal Bhatta, Sangeetha Kizhakkekalam, Navaneeth P K, Alex Markowitz, Zhicheng He, Santanu Mondal
Computational Engineering
From language-model stock rankings to testable economic rules: A computational audit
Shuai Wu, Xue Li, Zhijun Wang et al. (+4 more)
We test the stability, reproducibility and investment outcomes of
language-model stock rankings. Four models and five numerical comparators share a portfolio engine over 72 monthly holding periods in the Shanghai Stock Exchange (SSE) 50, China Securities Index (CSI) 300 and CSI 500. Rankings use nine characteristics, and five repeated SSE 50 runs measure variation under identical inputs. Linear rules fitted to development-period model preferences are frozen before unseen-month, larger-pool and controlled-intervention tests. Their mean Spearman agreement with model rankings is 0.923-0.984 in the SSE 50 and 0.795-0.985 after transfer. Aggregate rank-change error falls relative to a zero-change prediction in twelve archived feature-group comparisons and eight matched single-feature comparisons, with Holm adjustments applied in separate nine-plus-three and six-plus-two families. Prediction of individual entries and exits remains weak (event Jaccard 0.000-0.125). Historical mean model compound annual growth rates range from 6.26% to 11.17%. At 10 basis points per side and six-month blocks, the twelve-comparison model-minus-rule return family and factor-controlled associations yield no adjusted finding. Three higher-cost, twelve-month-block comparisons favor a Terra rule within their twelve-test slices, with no adjusted finding across the full 144-test sensitivity grid. Two input-intervention batches totaling 5,184 responses supply paired intervention-return tests. The three-model batch has no bootstrap-adjusted finding at the primary block length; a Luna row-order effect appears under heteroskedasticity- and autocorrelation-consistent (HAC) adjustment within its three-test family but not in a pooled 24-test adjustment. Compact rules approximate aggregate rankings; return conclusions depend on comparison families, uncertainty methods and tie priorities.
Read the full abstract ▾
All authors Shuai Wu, Xue Li, Zhijun Wang, Bolun Liu, Weilin Cai, Zihao Su, Ran Wang
Optics
Vector solitons in nonlinear lattices with competing SPM and XPM terms
Liangwei Zeng, Boris A. Malomed, Dumitru Mihalache et al. (+1 more)
We demonstrate that two-component photonic nonlinear
lattices with opposite signs of the out-of-phase spatially modulated SPM and XPM coefficients can support two types of optical vector
solitons (VSs), the fundamental and mixed ones. The fundamental VSs are composed of two Gaussian-like components, while the mixed ones combine Gaussian-like and dipole components. The structures and stability of both types of the VSs are investigated. The VSs are stable, except for narrow intervals at small values of the propagation constant.
Read the full abstract ▾
All authors Liangwei Zeng, Boris A. Malomed, Dumitru Mihalache, Xing Zhu
Astrophysics of Galaxies
Chemical complexity in feedback from supernova remnants: first detection of PO+ in IC443 and W44
G. Cosentino, F. Priestley, S. Scibelli et al. (+1 more)
Supernova remnant (SNR) may shape the physical and chemical properties of
star formation. By releasing refractory elements into the gas phase, they inject material of pre-biotic relevance into molecular clouds. Such elements may be inherited by planetary systems and become relevant for the emergence of life. We investigate the presence of P-bearing species in molecular clouds impacted by SNR-shocks. P is a crucial biological element, but its presence and reservoir in molecular clouds remain elusive. We present high-sensitivity 3 mm observations obtained with the IRAM 30m, toward the interaction sites between the SNRs W44 and IC443 and their associated molecular clouds. Focusing on PO+(2-1) and PN(2-1), we derive molecular column densities and abundances in local thermodynamic
equilibrium. We investigate the correlation between P-bearing species and the shock tracer Silicon Monoxide (SiO). We detect, for the first time, PO+ toward SNRs, with abundances X(PO+)~8e-12-5e-11, but no significant PN emission. PO+ toward both SNRs is associated with strong SiO emission, showing similar linewidths but velocity offsets, suggesting that PO+ preferentially traces the most compressed and more strongly ionised layers of the shock. By comparing the derived abundances with those reported in literature, we suggest that the PO+ abundance mainly depends on a combination of shock strength and shocked medium density. Our results indicate that P-chemistry in SNR shocks is shifted toward an ion-dominated regime, likely driven by enhanced
cosmic-ray ionisation rates. In this scenario, neutral species such as PN are efficiently destroyed, while ion-molecule reactions favour the production of PO+. This suggests that SNR shocks may play a role in processing P-bearing material and shaping gas-phase P chemistry in molecular clouds, with potential implications for the chemical inventory of
star-forming environments.
Read the full abstract ▾
All authors G. Cosentino, F. Priestley, S. Scibelli, F. Fontani
Computer Vision
EgoFound3R: End-to-End Egocentric Hand Reconstruction in World Space with Point-Wise Interaction Attributes
Hongming Fu, Jingcheng Shi, Wenjia Wang et al. (+2 more)
Egocentric video has become a primary source of supervision for
embodied models, and its value rests on recovering hand motion in world coordinates, which camera motion and hand occlusion make difficult. Existing reconstruction pipelines typically separate hand and scene estimation, leave interaction attributes to separate task-specific models, and invoke several models per
video, so no prior reconstruction model estimates these attributes and throughput becomes a practical constraint on large-scale annotation. We therefore introduce EgoFound3R, a unified end-to-end model that estimates world-space hand geometry in a metric scale shared with the scene, and predicts point-wise interaction attributes, including visibility, contact, and distance. The model integrates three designs: (i) structured hand
prompts that transfer pretrained
geometric priors to world-space hand reconstruction; (ii) an explicit hand representation that decodes hand geometry and interaction attributes; and (iii) a shared-parameter multi-rate design that lowers inference cost. Together, these designs predict hand geometry and point-wise attributes in one pass. On OakInk-v2, TACO, and HOI4D, EgoFound3R reduces the mean per-joint position error (MPJPE) by 43.2%, 22.4%, and 11.6% over previous methods and predicts point-wise contact and distance alongside the geometry in the same pass, while attaining approximately 6x higher throughput.
Read the full abstract ▾
All authors Hongming Fu, Jingcheng Shi, Wenjia Wang, Binhua Zuo, Bo Zhao
Quantum Physics
Fermionic Gaussian Scrooge Ensembles in Deep Thermalization
Ning Sun and Pengfei Zhang
Measuring part of a many-body wave function generates a projected ensemble of pure
quantum states on the unmeasured subsystem. Recent advances in deep thermalization have shown that, in chaotic systems, this ensemble universally converges to the maximally random ensemble compatible with its average density
matrix, known as the Scrooge ensemble. By contrast, free-
fermion systems are nonchaotic, and their
quantum states are constrained to remain Gaussian. Motivated by this distinction, we introduce the fermionic Gaussian Scrooge ensemble to describe deep thermalization in generic free-
fermion systems. This ensemble is defined as a distortion of the Gaussian Haar ensemble by the average density
matrix, and its moments reveal an enlarged
symmetry of Gaussian states in the replicated
Hilbert space. We demonstrate the emergence of the fermionic Gaussian Scrooge ensemble in two complementary settings: (1) we analytically prove that the projected ensemble of the SYK$_2$ model follows the fermionic Gaussian Scrooge ensemble at arbitrary
evolution times; (2) we provide numerical evidence that it emerges at sufficiently long times in random Gaussian
circuits with charge conservation. Our results establish the fermionic Gaussian Scrooge ensemble as a universal description of deep thermalization in free-
fermion systems.
Read the full abstract ▾
Operator Algebras
The crystal groupoid of a compact semisimple Lie group and its flag varieties
Aidan Sims and Robert Yuncken
We introduce the
crystal groupoid of a compact semisimple Lie group $K$; it is the Deaconu-Renault groupoid of a higher-rank analogue of a subshift of finite type on an inverse limit of Kashiwara
crystals. We show that its Steinberg
algebra and its $C^*$-
algebra realise the
crystal limits of the
polynomial function
algebra of $K$ and its $C^*$-envelope, respectively, in the sense of Matassa and the second author. We show that the orbit space of the
crystal groupoid is in bijection with the Weyl group of $K$, with
topology determined by the Bruhat order on the Weyl group. We also define a
crystal groupoid for flag
manifolds and related Poisson homogeneous spaces, and prove the analogous structural results.
Read the full abstract ▾
Artificial Intelligence
Dependency-Aware Reward Shaping for Agentic Reinforcement Learning
Ziyi Chen, Yan Zhang, Jianhui Wei et al. (+2 more)
When training large
language models with reinforcement learning, terminal rewards provide little guidance about which steps matter. Common methods for assigning step credit overlook that work built on uncorrected mistakes is wasted while independent work remains valid. With only a final success/failure reward, every step in a failed episode has zero total future reward, even when it made progress. We propose Dependency-Aware Reward Shaping (DARS), which represents task progress as predicates linked by prerequisite relations and assigns step-level credit over the dependency graph. An annotator marks which predicates each step verifies, invalidates, or repairs. Verified predicates are discounted according to graph distance from the nearest broken prerequisite, while independent predicates are unaffected. Repairs update these weights based on any errors that remain; invalidated predicates need re-verification to regain credit. A fixed potential converts these annotations into signed per-step rewards. A common reward and annotation interface allows DARS to integrate with a range of
reasoning and agentic training methods, such as GiGPO and ARPO/AEPO, without changing their rollout strategies or optimizers. Across five task families and models from 1.5B to 8B, DARS improves success by up to 10 points over GiGPO trained with the same budget and harness (ALFWorld), raises the WebShop task score and Search-R1 QA accuracy, complements AEPO's
entropy-based training on AIME24/25 with a Python interpreter, and exceeds OmniOPD in controlled tool-free
reasoning comparisons at 1.7B and 4B. Ablations show that step-level credit, dependency attenuation, and graph
topology each contribute. On ALFWorld, a distilled 8B annotator matches the API annotator, enabling DARS to run efficiently without a frontier judge. Code is available at https://github.com/JianhuiWei7/DARS.
Read the full abstract ▾
All authors Ziyi Chen, Yan Zhang, Jianhui Wei, Daoan Zhang, Zuozhu Liu
Computer Vision
Resolving Mixed Single-Photon LiDAR Returns for Foreground-View and Hidden Scene Reconstruction
Ziting Wen, Runrong Deng, Zili Zhang et al. (+5 more)
Partially transmissive screens and protective covers are common in robotic inspection, but they create mixed
LiDAR returns from both the foreground material and the scene behind it. Conventional peak-based
LiDAR usually discards weak hidden returns, while single-
photon LiDAR records time-resolved histograms that preserve attenuated and overlapping echoes. However, existing transient reconstruction methods typically fit a single scene representation to the measured waveform. Under occlusion, weak or nearby foreground–hidden echoes can form a broad peak or subtle shoulder. Because such waveforms can also be explained by a displaced single surface or a thick density distribution, accurate transient fitting does not necessarily imply correct geometry. We propose a state-aware framework for foreground-view and hidden scene reconstruction from occluded single-
photon histograms. For each ray, we estimate local echo evidence, identifying no reliable surface evidence, single-return evidence, or two returns. The inferred echo state routes supervision for a two-head neural field: all rays constrain waveform reconstruction, while reliable anchors provide geometry localization. We also introduce a real paired single-
photon LiDAR occlusion
dataset with occluded and clean captures at fixed poses. Experiments on a real
dataset show improved hidden scene depth and point-cloud accuracy over baselines. Our results demonstrate single-
photon layered reconstruction as a practical route for 3D perception through partially transmissive occluders.
Read the full abstract ▾
All authors Ziting Wen, Runrong Deng, Zili Zhang, Haitao Zheng, Yuecong Xu, Xiaoqiang Ren, Guodong Shi, Kemi Ding
Computer Vision
Semantic RGB–Depth Based Surgical Skill Assessment in Microscopic Stereo Videos
Jecia Z. Y. Mao, Sue M. Cho, Francis X. Creighton et al. (+3 more)
Objective assessment of microsurgical technical skill is essential for competency-based training and quality assurance, yet existing
video-based approaches predominantly rely on RGB images and therefore overlook the 3D spatial relationships that characterize instrument-anatomy interactions. Although stereo operating microscopes provide complementary depth information, conventional stereo matching
algorithms can produce sparse and unreliable depth estimates under high-magnification imaging conditions, limiting their use for automated skill assessment. This work presents a semantic RGB-Depth framework for
surgical skill assessment from microscopic stereo
videos. A regression-based depth fusion method combines sparse metric stereo depth with dense
monocular depth estimates to generate a dense
geometric representation of the
surgical scene. This representation is integrated with semantically decomposed RGB streams corresponding to individual
surgical instruments and surrounding anatomy. A hierarchical
attention architecture jointly encodes these streams to capture discriminative patterns of instrument use and instrument-anatomy interaction across surgeons at different training levels. The framework was evaluated on 33 ex vivo transoral microlaryngeal procedures performed by six surgeons, comprising attending surgeons and
surgical residents, using leave-one-surgeon-out cross-validation. The proposed semantic RGB-Depth model achieved an F1 score of 0.938 for skill-level classification, compared with 0.696 for semantic RGB and 0.929 for semantic depth. These results suggest that
geometric information can improve automated
surgical skill assessment from microscopic stereo
videos. The learned spatial, temporal, and semantic
attention patterns also support qualitative examination of the scene regions,
video segments, and semantic streams emphasized by the model.
Read the full abstract ▾
All authors Jecia Z. Y. Mao, Sue M. Cho, Francis X. Creighton, Deepa Galaiya, Russell H. Taylor, Manish Sahu
Machine Learning
Autoregressive Drillhole Modelling Under Distribution Shift
Yihao Ding, Daniel Yitian Su, Yiran Zhang et al. (+2 more)
Autoregressive modelling has achieved remarkable success in
language and sequence tasks by learning to predict future states from previous observation. Mineral-exploration drillholes provide a natural but largely unexplored setting for this paradigm: as drilling proceeds, lithology is revealed sequentially from shallow to deep, making prediction of deeper strata inherently autoregressive. Existing drillhole modelling, however, is dominated by spatial interpolation and reconstruction, or largely rely on masked modelling, leaving strictly autoregressive prediction largely underexplored. We introduce DrillBench, a
benchmark of 49,671 Western Australian drillholes for next-layer prediction and autoregressive stratigraphic generation across a graded transfer spectrum, from local prediction through spatial shift to cross geological province transfer. Benchmarking classical, geostatistical, and neural models reveals a clear transfer boundary: spatial and geochemical conditioning provides large local gains but deteriorates sharply under stronger shift, whereas lithology-sequence autoregressive models transfer more robustly. Guided by this finding, we develop a backbone-agnostic recipe combining large-scale pretraining on historical drillholes with spatial
retrieval of neighbouring lithology.
Retrieval is most effective in weathered cover, when local spatial continuity remains informative, whereas pretraining contributes more strongly in bedrock and under broader geological shift. Together, they retain strong local performance while improving generalisation under spatial and cross-province shift, most markedly on the most distant splits. The
benchmark and code are available at https://github.com/yihaoding/drillbench.
Read the full abstract ▾
All authors Yihao Ding, Daniel Yitian Su, Yiran Zhang, Christopher M. Gonzalez, Wei Liu
Software Engineering
CompProv Produces Machine Readable Graphs Encoding Microscopic Algebraic Provenance for Reproducible Computation
Minas Abramyan, Mohammed Alaa Ala'anzy and Nasir Saeed
The reliability of computational results in scientific research and financial modeling increasingly depends on verifiable traceability, not merely on trust in a reported output. Existing provenance systems operate at the granularity of files,
datasets, or pipeline stages, leaving the internal sequence of algebraic transformations connecting an
algorithm's inputs to its outputs unrecorded; a rounding error, an undocumented substitution, or a missing intermediate value can propagate to a final result with no recoverable trace. This work presents CompProv, a Java-based, audit-oriented provenance framework that captures lineage at the atomic level of individual algebraic operations by encapsulating numerical values in high-precision wrapper objects, producing a serializable Calculation Provenance Graph (CPG) that persists as a self-contained artifact rather than a discarded byproduct. The framework is evaluated through three heterogeneous case studies: a decentralized-finance NAV calculation, a reconstruction of an interferometric gauge-block calibration in metrology under explicitly documented input assumptions, and a hydrological model performance evaluation. Deterministic replay reproduced each result exactly in a fresh environment, and CPG-based input substitution supported sensitivity analysis without exposing the underlying source code. These results indicate that, provided the CompProv runtime and wrapper classes are available in the replay environment, a CPG allows numerical integrity to be audited without disclosing proprietary business logic, and that its self-contained structure supports temporal auditability once an execution environment has become deprecated. This framework establishes an operation-level foundation for auditable-by-design computational systems, with scaling to high-throughput computing identified as a direction for future work.
Read the full abstract ▾
Differential Geometry
Parabolic Monge-Ampère equations on a family of Calabi-Yau manifolds
Jiyuan Han
We prove a uniform $L^\infty$-estimate for the
parabolic Monge–Ampère equation under uniform Hörmander-Tian type estimates with respect to the initial and prescribed measures. We apply this estimate to rescaled Kähler–
Ricci flows on a family of polarized Calabi–Yau
manifolds over the punctured disk. An interpolation argument gives uniform Hörmander-Tian type estimates along the flows. We also establish uniform diameter bounds and Gromov–Hausdorff precompactness for this family of flows. In particular, we obtain an alternative proof of the diameter estimate by Li-Tosatti.
Read the full abstract ▾
Computer Vision
iSEE: Object Permanence Through Self-Supervision
Pramish Paudel, Ajad Chhatkuli, Luc Van Gool et al. (+1 more)
Object permanence, keeping track of an object's identity and position while it is occluded, is central to
video representations that track, predict and plan. Trackers that achieve it learn from boxes, track identities and visibility labels. On the other hand, self-supervised object-centric methods discover objects without labels: through slot
attention, it represents a
video as slots that bind to objects and follow them across frames. However, these slots are lost under occlusion, making the desired permanence impossible.
Reasoning permanence is a hard problem because it requires to detect when an object becomes occluded, re-identify when object reappears, and keep the object's hidden position continuous, using reapperance as the only learning cue. To address this, we propose iSEE, a novel framework that offers all three aforementioned requirements, without any labels whatsoever. We built iSEE using the following three proposed components: (i) Object evidence modelling: a slot's
attention, compared with its own past, reveals when its object is hidden. (ii) Appearance-position separation: two slot streams let the appearance be held for re-identification while the position keeps changing. (iii) Permanence from reappearance: a walker follows the hidden object's position, trained only on where the object reappears. On LA-CATER static, iSEE returns a reappearing object to its own slot after 86% of occlusions, against 32% for SlotContrast, and localises it while hidden within 4.1 mAP of the label-trained SoTA RAM. The two streams also allow downstream planning, with the position stream as the action of a world model. Project page: https://insait-institute.github.io/iSEE/
Read the full abstract ▾
All authors Pramish Paudel, Ajad Chhatkuli, Luc Van Gool, Danda Pani Paudel
Mesoscale and Nanoscale Physics
Phonon interference induced by defect pairs
Ran-Bo Yang, Ming-Hao Li, Chun-Yu Cai et al. (+4 more)
The coherent dynamics of
lattice waves is of significant importance for exploring the nonadiabatic phenomena of the electron-
phonon coupling under nonequilibrium conditions. Here, we propose a theoretical model for
phonon interference induced by a defect pair, in which
lattice waves with the selective modes are launched independently by each of defects due to the strongly local electron-
phonon coupling. We find that the intensities of
phonon interference fringes could be directly evaluated by a measurable quantity in experiments–Huang-Rhys factor, and the geometry of interference patterns could be modified by the internal parameters of defect pair, including the charge states, spatial distance, energy-level depth and localization strength. In particular, several novel
quantum states, such as the Dirac-like and nodal-line topological states are emerged for the appropriate internal parameters. Furthermore, using Huang-Rhys model, we simulate the ultrafast optical spectroscopy in the
phonon interference duration when defect pairs are excited as the color centers, where interference fringes are perfectly matched by intensities of spectrum, similar to the $\textit{lock-key}$ pattern. These results show that defect pair, serving as the driven force for the coherent dynamical processes for multidegrees of freedom, offers an ideal platform to explore fundamental
quantum phenomena in nonequilibrium and ultrafast physics.
Read the full abstract ▾
All authors Ran-Bo Yang, Ming-Hao Li, Chun-Yu Cai, Yue-Wen Wang, Zhi-Qing Li, Li-Xia Zhao, Zi-Wu Wang
Machine Learning
Low-Budget Active Learning through Entropic Optimal Transport
Rim Hajal, Mathieu Besançon and Jérôme Malick
We consider low-budget active learning, which consists of selecting a limited number of points, the coreset, such that a model can be trained to high accuracy on the selection only. This problem is particularly relevant in contexts where labeling requires costly expert intervention, as in medical applications. We leverage features extracted from a pretrained self-supervised model to represent the data, and perform coreset selection directly in this feature space. In this paper, we use entropic optimal transport, specifically the Sinkhorn divergence, as the coreset selection criterion, which first allows us to get dimension-free sample complexity results, and second admits computationally efficient gradient evaluations. This opens the way to using gradient-based
algorithms to rapidly compute solution candidates, further improved by a swap-based local search, with guarantees on the solution quality. Experiments on image
benchmarks and medical
datasets show that our method outperforms state-of-the-art heuristics in low-budget settings.
Read the full abstract ▾
Programming Languages
Associativity and Commutativity in Equality Saturation
Tarik Rosin, Marcel Ullrich and Sebastian Hack
Equality saturation is a promising technique for program optimization which sidesteps the phase ordering problem. However, current e-graph implementations grow exponentially large, even for simple examples. Many practical applications involve associative and commutative (AC) operators. We present an extension of relational e-matching that handles AC operators natively by storing terms as multisets. Preliminary results show that equality saturation modulo AC uses asymptotically less
memory in certain cases.
Read the full abstract ▾
Optimization and Control
On solving integer bilevel optimization problems with a non-convex quadratic follower objective function using disjunctive cuts
Kübra Tan{\i}nm{\i}ş, Elisabeth Gaar, Jon Lee et al. (+2 more)
In this work, we study bilevel optimization problems where all variables are integer, all constraints and the leader objective function are linear, and the follower objective function is non-convex quadratic. Relying on bilevel-free sets derived from improving directions, we develop a disjunctive cut approach to exclude bilevel-infeasible solutions within a branch-and-cut
algorithm. We show that our disjunctive cuts can be obtained by solving a cut generating linear program. Furthermore, we discuss conditions that allow the number of disjuncts in the cut generating linear program to be reduced, and we propose several strategies to identify improving directions and generate disjunctive cuts efficiently. We evaluate various aspects of the proposed branch-and-cut
algorithm on both convex instances from the literature that fit our setting and new non-convex instances and compare the performance of our best approach with existing state-of-the-art approaches, which we significantly outperform.
Read the full abstract ▾
All authors Kübra Tan{\i}nm{\i}ş, Elisabeth Gaar, Jon Lee, Ivana Ljubić, Markus Sinnl
Dynamical Systems
Asymptotic Expansive and Hyperbolic Linear Operators
Priyabrata Bag, Pabitra Narayan Mandal and Pramod Kumar Das
We study the topological dynamical notion of asymptotic expansivity for linear operators. We also introduce two new notions, namely strong asymptotic expansivity and super asymptotic expansivity. We show that all three notions are equivalent to hyperbolicity for linear operators on finite dimensional spaces. In case of infinite-dimension, we construct examples to show that asymptotic expansivity is weaker than strong asymptotic expansivity which is further weaker than super asymptotic expansivity. More interestingly, we prove that super asymptotic expansivity is equivalent to the notion of hyperbolicity. Then, we prove that the super asymptotically expansive linear operators is a dense class in strong operator
topology. Finally, we show elegant ways to construct variants of asymptotically expansive linear operators.
Read the full abstract ▾
Artificial Intelligence
Qiang Yang, Zhiqiang Kou, Xueyi Zhang et al. (+6 more)
Large
language model (LLM)
agents increasingly operate in private environments and accumulate valuable experience from task execution, tool use, feedback, and local knowledge. Yet such experience is distributed across organizations and cannot be directly shared because of
privacy and proprietary constraints. Conventional
federated learning is insufficient for this setting, as
agent capabilities extend beyond model parameters to
memory, tools, rewards, skills, and structured knowledge. In this paper, we formulate
Federated Agent Optimization (FAO), which studies how distributed
agents can collaboratively improve through controlled information exchange while keeping raw data, complete trajectories, and private knowledge local. We define FAO as a multi-objective problem balancing
agent utility,
privacy leakage, and communication cost, and organize its optimization space across policy,
memory, tool use, reward, and structured knowledge and skills. We further characterize how private experience can be abstracted, protected, aggregated, and adapted into transferable capabilities, providing a unified view of how
agents can benefit from one another without direct experience sharing. Finally, we identify the key challenges of FAO and outline several promising directions for future research toward trustworthy
federated agent systems.
Read the full abstract ▾
All authors Qiang Yang, Zhiqiang Kou, Xueyi Zhang, Dong-Dong Wu, Hanlin Gu, Jing Guo, Yang Liu, Di Jiang, Qian Xu
Mesoscale and Nanoscale Physics
Synthesis of Metal-Semiconductor Heterojunctions in Single Graphene Nanoribbons
Ziyi Wang, Kaitlin Slicker, Weichen Tang et al. (+9 more)
Graphene nanoribbons (GNRs) are a highly tunable class of one-dimensional (1D)
quantum materials that can be fabricated through bottom-up synthesis. Precise control over GNR band gaps and band-edge
alignments has established them as a promising nanoelectronics platform, but forming high-quality electronic interfaces remains challenging. Here we combine sequential on-surface synthesis and scanning tunneling microscopy (STM)-induced dehydrogenation to directly write metallic 7-iGNR segments into otherwise semiconducting H$_2$-7-iGNRs without changing the GNR width or carbon backbone connectivity. The resulting metallic segments exhibit nonzero spectral weight at the
Fermi level ($E_\mathrm{F}$) and spatially extended electronic states, consistent with two dispersive bands crossing $E_\mathrm{F}$. The new metallic states are well described by an extended Su–Schrieffer–Heeger zigzag-ladder
Hamiltonian. Scanning tunnelling spectroscopy (STS) and first-principles calculations show that the valence band edge of adjacent, connected semiconducting GNR segments lies close to $E_\mathrm{F}$ for the metallic segments. STS spectra recover bulk GNR properties within $\sim$1 nm on either side of the metal-
semiconductor GNR interface, with no discernible depletion region. The observed GNR band edge
alignment is consistent with a small energy barrier for hole injection, thus suggesting a possible route towards atomically precise, low-barrier p-type contacts for future GNR-based electronic devices.
Read the full abstract ▾
All authors Ziyi Wang, Kaitlin Slicker, Weichen Tang, Boyu Qie, Rafal Zuzak, Haowen Pang, Yudi Huang, Xinheng Li, Peter H. Jacobse, Steven G. Louie, Felix R. Fischer, Michael F. Crommie
Machine Learning
Counterfactual Generation via Flow Matching: Coupling-Sensitive End-to-End Rates
Yunrui Guan, Krishnakumar Balasubramanian and Shiva Prasad Kasiviswanathan
Counterfactual generation seeks to sample outcomes under a hypothetical intervention or decision using observational data collected under the factual assignment mechanism. We develop a flow-matching approach that combines a sample-split, doubly robust training objective with a learned coupling between observed source outcomes and target outcomes drawn from a fitted conditional outcome model. To enable finite-step generation, we leverage a score-corrected
stochastic sampler based on a Gaussian-smoothed interpolation. Our main theoretical contribution is a coupling-sensitive KL bound for constant-step Euler discretization: the error is controlled by moments of the source–target displacement under the chosen coupling, rather than by global uniform regularity of the velocity field, and has near-linear dependence on the ambient dimension. We also establish finite-sample non-parametric guarantees for the learned velocity and score fields when both the conditional outcome model and the source-target coupling are estimated from data. These bounds separate approximation, coupling-replacement, nuisance-estimation, generalization, and Monte Carlo errors and, combined with the sampler analysis, yield an end-to-end guarantee for counterfactual generation. Experiments on synthetic and semi-synthetic image
benchmarks support the coupling-dependent theory and show that, at finite discretization budgets, the
stochastic sampler can outperform the corresponding deterministic ODE sampler.
Read the full abstract ▾
Computer Vision
FlashBack: Knowing When to Remember in Streaming Vision-Language Models
Yi Chen, MingMing Yu, Rui-Qi Wang et al. (+5 more)
Streaming
vision-
language models must process continuously growing
video streams under a bounded compute budget, creating a persistent tension between real-time perception and long-term
memory. Retrieving historical information provides a natural remedy, yet historical recall is not uniformly beneficial: unnecessary history may introduce irrelevant context into current
reasoning and interfere with native real-time perception. Effective streaming
memory should therefore address not only what to remember, but also when and how to access it. To this end, we introduce FlashBack, a training-free framework for selective, multi-level
memory in streaming
vision-
language models. Before retrieving history, FlashBack draws on the semantic understanding of the frozen streaming VLM to infer whether a query calls for historical evidence. This assessment determines whether inference remains on the Native trajectory or invokes an isolated Recall trajectory. The Recall trajectory combines recent context with retrieved long-term
memory through a query-local Side-KV pathway, preserving local temporal continuity without modifying the persistent Native state. We instantiate FlashBack on StreamingVLM and Mage-VL-4B and evaluate it on OVO-Bench and StreamingBench. The results show improvements on several long-horizon and
memory-dependent tasks while largely preserving real-time perception, with performance competitive with strong training-based streaming methods despite requiring no additional training. Our code will be announced later.
Read the full abstract ▾
All authors Yi Chen, MingMing Yu, Rui-Qi Wang, Boran Wang, Xiaohang Cao, Chu Tang, Jingmin Chen, Jie Gu
Computer Vision
Color Independent Word Segmentation From Transcribed Bangla Passages
Faias Satter, Noor Masrur and Sk. Md. Masudul Ahsan
An optical character recognition(OCR) system can scan paper and extract text, making people's jobs easier. While numerous OCR systems are accessible in the software sector, finding a dependable equivalent solution for Bangla is tough. When it comes to handwritten texts, the case is even more rare. The first fundamental step to any OCR is to segment words from text images. If this stage fails, the total OCR's performance will be poor no matter how promising the later stages perform. This research aims to segment words in a handwritten Bangla text image. This research can be implemented on any smartphone-captured image, irrespective of the color and type of paper and ink. Furthermore, as smartphone-captured images can create shadow interferences, the custom
dataset built for this research is created in such a way that every possible obstacle that can be faced is included. For 7374 words, a total of 7278 bounding boxes are generated, which have recall of 90.60%, precision of 91.80%, and F1-score of 91.20%. The system can be further improved with nested operations on bounding boxes containing several words or by adjusting the adaptive thresholding and dilation filter sizes to a more precise level.
Read the full abstract ▾
Quantum Physics
Rapid mixing of quantum spin chains at any finite temperature
Leeseok Kim
We prove that a quasi-local
quantum Gibbs sampler rapidly mixes to the Gibbs state of any 1D finite-range
Hamiltonian at all finite temperatures in $\mathcal{O}\left(\log (n/\varepsilon)\right)$ time. Each update traces out a contiguous block of constant length and reconstructs it using the Petz recovery map, following early proposals for
quantum heat-bath Gibbs samplers by Kastoryano and Brandão~\cite{kastoryano2016
quantum}. Our proof directly bounds the distance between the updated state and the target Gibbs state using
quantum Wasserstein distance of order $1$. We further provide $\mathcal{O}(\text{polylog}(n/\varepsilon))$ depth of
quantum circuits to prepare these Gibbs states. We hope these techniques will facilitate the analysis and design of other Gibbs samplers.
Read the full abstract ▾