Searcharxiv⌕ Search

arXiv subjects

Zhihao Zhao

Publications and source records attributed to Zhihao Zhao.

At least 19 recordsLinked to original sources

Early Readability of Trajectory Quality in Diffusion Driving Planners

Diffusion-based driving planners can generate multiple trajectories for the same scene, but these candidates vary in quality. Can their internal states reveal which candidate will be better before denoising is complete? We systematically study this question in three pretrained planners: Alpamayo, ReCogDrive-IL, and HDP, using linear probes to predict final trajectory quality from intermediate states and evaluating candidate ranking and selection on held-out scenes. Across all three planners, we observe that states from the first denoiser evaluation already support informative selection, while further denoising does not consistently improve selection quality. Further analyses show that readout depends on the representation and reader, and that selection can coexist with limited overall ranking accuracy. These findings motivate an early-selection method that selects after the first denoiser evaluation and completes only the chosen candidate. This reduces computation relative to fully denoising every candidate and improves open-loop trajectory quality over random selection across all three planners. In particular, it reduces Alpamayo's average displacement error by 17.8% and 13.4% in one- and four-camera configurations, respectively. Closed-loop evaluations also show higher mean driving scores and progress.

cs.RO↗

Regular integral models of Shimura varieties and a conjecture of Pappas

We construct regular integral models for a class of Shimura varieties of symplectic and orthogonal type with maximal parahoric level at an odd prime $p$. These models are defined over the ring of integers of the reflex field and have special fiber with normal crossings and irreducible components of multiplicity one or two. Our construction uses explicit equivariant modifications of the corresponding canonical local models for symplectic and split even orthogonal similitude groups, with the minuscule cocharacters corresponding to maximal isotropic Grassmannians. These modifications are obtained by successively blowing up Schubert varieties in the special fiber. This proves the equivariant modification conjecture of Pappas at maximal parahoric level in the above cases.

math.NT↗

Closing the realism gap in physics-based gait simulations with a learned state prior

Predictive simulation of human movement is a promising tool for studying ``what-if'' scenarios in human movement and its underlying motor control, yet its realism is often limited. To address this gap, we incorporate a learned state prior that is trained on a large-scale dataset of human gait kinematics and external forces into predictive simulations. Resulting gait simulations yield kinematics and kinetics across diverse walking and running speeds that better match experimental data than current physics-based simulations, achieving accuracy comparable to data-driven models that reproduce learned data. Furthermore, our method enables robust hypothesis testing by demonstrating how varying optimality assumptions, muscle weakness, and footwear choices influence predicted gait. We also show that this prior generalizes well beyond its training data, successfully reconstructing full-body kinematics for curved running and cutting maneuvers from sparse marker sets. Ultimately, these results suggest that state priors should be broadly integrated into predictive simulations.

cs.CE↗

Longitudinal Retinal Vascular Remodeling in Myopic Children Treated with Orthokeratology or Defocus Lenses: A Two-Year Comparative Study

Purposes: To characterize longitudinal retinal vascular changes in myopic children treated with orthokeratology (OK) or multifocal defocus lenses (Defocus) and to examine their association with axial elongation. Methods: In this retrospective cohort study, 43 myopic children underwent comprehensive clinical examination and fundus photography at baseline, 12 months, and 24 months. Axial length (AL) and spherical equivalent refraction (SER) were recorded at baseline, 6, 12, and 24 months. An automated segmentation model extracted vascular parameters, main vessel angle (MA), branching angle (BA), bifurcation edge angle (BEA), crossover point (COP), and terminal vessel count (TVC). Repeated-measures ANOVA assessed temporal changes. Pearson or Spearman correlations evaluated associations between AL and vascular metrics. Results: Over 24 months, the OK group exhibited significantly slower axial elongation than the Defocus group (0.214 mm and 0.522 mm, p < 0.01). In the OK group, MA and BA decreased modestly, BEA in arteries declined gradually, but COP and TVC remained relatively stable. The Defocus group demonstrated more pronounced decreases in MA and BA, an increase in BEA, and significant reductions in COP and TVC (p < 0.05). Correlation analysis revealed stronger associations between AL and vascular parameters, especially COP and TVC, in the Defocus group at all time points, whereas only BA and BEA correlated with AL in the OK group. Conclusions: OK lenses mitigate axial elongation and induce milder retinal vascular remodeling compared to Defocus lenses. Distinct temporal patterns of vascular metrics changes were observed between the two interventions, and correlate differentially with axial growth.

cs.CV↗

Noise-intensity bifurcations of transition paths by Morse index formula

We study most probable transition paths (MPTPs) of a gradient diffusion system in $\mathbb{R}^n$ with fixed endpoints, in the framework of Onsager--Machlup (OM) theory. The existence theory for such paths is developed within Lagrangian variational theory: above the Mañé critical value $c_u(L)$ the energy-penalized action admits global minimizers, while below the endpoint-dependent critical value $k_0(L;x_\pm)$ only time-truncated minimizers exist; at regular levels between them, free-time extremals exist. The main result is a $\{0,1\}$-index theorem: at a nondegenerate free-time extremal, the Morse index of the full Hessian equals the fixed-time Morse index plus a correction $ν\in\{0,1\}$, equal to $1$ exactly when the reduced (minimal fixed-time) action satisfies $S''(T)<0$. The loss of local minimality along a branch is thereby split into two mechanisms: a conjugate-point mechanism and a duration mechanism. Combined with a Hamiltonian spectral-flow formula, this yields a bifurcation criterion in the noise intensity $σ$ and a sharp local stability criterion expressed in terms of the crossing instants. As an example, the one-dimensional quartic double well is analyzed in detail.

math.DS↗

Physical Kernel: Structured Visual Latents for Dark Manipulation

We study dark manipulation: after a brief lit Write encodes z0 = Enc(rgb), a policy pi(z) and open-loop dynamics f(z,a) complete contact-rich skills without further pixels (dark_f). On ManiSkill StackCube (n=160; seed packs 0/1000), dark_f attains 68.1% stacked on the five-rung chain (near_A -> grasped -> lifted -> on_B -> stacked), compared with 35.6% for per-step lit_reenc and 0% for freeze/encode_black. On a shared Write->HOLD protocol (n=40), occlusion and camera-aligned GT contact-neighbor masks drive lit lift from 43% to 0%, while dark_f holds 82.5%; shuffling actions inflates dynamics MSE by ~9.4x; write-time appearance shifts break encoding (night: 0% stacked), yet the same shifts during HOLD leave dark_f lift unchanged; Write length Tw is flat once the stop phase is reached, while earlier stops and write-time blur/JPEG sharply cut stacked. A dedicated pi_write reaches 35% vision-budget stacked (n=80); matched Dreamer-style/pixel nulls without privileged geom stay at 0%. Privileged state-RSSM MPC reaches ~35% stacked with 9D dark observations -- a stronger-observation null, not a matched visual baseline.

cs.RO↗

Automated 2D and 3D Segmentation of AMD and DME Lesions in OCT

Age-related macular degeneration (AMD) and diabetic macular edema (DME) are leading causes of vision loss, and optical coherence tomography (OCT) is the standard modality for detecting and monitoring the subtle lesions that drive treatment decisions. Most deep-learning segmentation work for OCT is validated only in-domain, leaving generalization to clinical data collected under different acquisition protocols largely untested. This work develops and systematically ablates four lesion-segmentation pipelines -- 2D and 3D variants for AMD and DME -- reaching Dice scores of 0.76 to 0.82 with strong volumetric and surface calibration (r vol, r surf greater than or equal to 0.97 across all four pipelines) on an in-domain validation set. The ablation process establishes a full-volume, calibration-aware adoption standard that catches mechanisms an ordinary slice-level evaluation would keep, and identifies ensemble composition as the most consistent driver of improvement. To test generalization, the models are evaluated on OLIVES, an external clinical cohort with no lesion-level ground truth, using a proxy-metric framework built around biomarker AUROC, central subfield thickness (CST) correlation, and longitudinal concordance. Predictions track clinical biomarkers outside the training distribution, though less strongly than in-domain -- evidence for, not validation of, automated lesion-burden tracking as a clinical tool.

cs.CV↗

On semi-stable integral models for Shimura varieties

We construct (potentially) semi-stable integral models for a class of Shimura varieties with maximal parahoric level at an odd prime $p$. The main input is an explicit construction of semi-stable equivariant modifications of the relevant canonical local models, where the underlying groups are Weil restrictions of unramified unitary similitude groups, symplectic similitude groups, or even orthogonal similitude groups. Via the local model diagram, these constructions give (potentially) semi-stable integral models of the corresponding Shimura varieties. In particular, the resulting models are regular and have reduced special fiber with normal crossings. As an application, we deduce the unipotence of the inertia action on nearby cycles and the $\ell$-adic cohomology of the geometric generic fibers.

math.NT↗

On $p$-adic integral moduli schemes and local models for PEL type D

We construct flat integral moduli schemes of PEL type D and the corresponding flat orthogonal Rapoport--Zink spaces with parahoric level structure over a $p$-adic integer ring. The construction relies on proving a conjecture of Pappas--Rapoport: for an even orthogonal similitude group over a complete discretely valued field of residue characteristic $p>2$, and for arbitrary parahoric level, the associated spin local model is flat, normal, Cohen--Macaulay, with reduced special fiber. In the course of the proof, we also show that in the quasi-split but non-split case, the Rapoport--Zink (naive) local model is topologically flat, verifying a conjecture of Pappas--Rapoport--Smithling. In the maximal parahoric case, we also describe the Schubert varieties in the special fiber in moduli-theoretic terms. Finally, for a maximal parahoric case we construct an explicit regular semi-stable model by blowing up the spin local model along the unique closed Schubert cell in its special fiber.

math.NT↗

KaLDeX: Kalman Filter based Linear Deformable Cross Attention for Retina Vessel Segmentation

Background and Objective: In the realm of ophthalmic imaging, accurate vascular segmentation is paramount for diagnosing and managing various eye diseases. Contemporary deep learning-based vascular segmentation models rival human accuracy but still face substantial challenges in accurately segmenting minuscule blood vessels in neural network applications. Due to the necessity of multiple downsampling operations in the CNN models, fine details from high-resolution images are inevitably lost. The objective of this study is to design a structure to capture the delicate and small blood vessels. Methods: To address these issues, we propose a novel network (KaLDeX) for vascular segmentation leveraging a Kalman filter based linear deformable cross attention (LDCA) module, integrated within a UNet++ framework. Our approach is based on two key components: Kalman filter (KF) based linear deformable convolution (LD) and cross-attention (CA) modules. The LD module is designed to adaptively adjust the focus on thin vessels that might be overlooked in standard convolution. The CA module improves the global understanding of vascular structures by aggregating the detailed features from the LD module with the high level features from the UNet++ architecture. Finally, we adopt a topological loss function based on persistent homology to constrain the topological continuity of the segmentation. Results: The proposed method is evaluated on retinal fundus image datasets (DRIVE, CHASE_BD1, and STARE) as well as the 3mm and 6mm of the OCTA-500 dataset, achieving an average accuracy (ACC) of 97.25%, 97.77%, 97.85%, 98.89%, and 98.21%, respectively. Conclusions: Empirical evidence shows that our method outperforms the current best models on different vessel segmentation datasets. Our source code is available at: https://github.com/AIEyeSystem/KalDeX.

eess.IV↗

The basic locus of regular ramified unitary Rapoport-Zink spaces at vertex-stabilizer level

We construct the Bruhat-Tits stratification of the reduced basic locus of regular ramified unitary Rapoport-Zink spaces of signature $(n\!-\!1,1)$ at vertex-stabilizer level. To study the Bruhat-Tits strata, we introduce strata models$\!-\!$simpler models that are étale-locally isomorphic to each stratum. They admit two complementary characterizations: (i) as strict transforms under the blow-up of the local model at its worst point, and (ii) via a partial moduli description given by explicit linear-algebraic conditions; from these we deduce smoothness, explicit dimension formulas and irreducibility of the Bruhat-Tits strata.

math.NT↗

The basic locus of unitary splitting Rapoport-Zink spaces with vertex stabilizer level

We construct the Bruhat-Tits stratification of the ramified unitary splitting Rapoport-Zink space, with the level being the stabilizer of a vertex lattice. To determine certain local properties of the Bruhat-Tits strata, we develop a theory of the strata splitting models. To study their global structure, we establish an explicit isomorphism between the Bruhat-Tits strata and certain (modified) Deligne-Lusztig varieties.

math.NT↗

Dynamic Structural Recovery Parameters Enhance Prediction of Visual Outcomes After Macular Hole Surgery

Purpose: To introduce novel dynamic structural parameters and evaluate their integration within a multimodal deep learning (DL) framework for predicting postoperative visual recovery in idiopathic full-thickness macular hole (iFTMH) patients. Methods: We utilized a publicly available longitudinal OCT dataset at five stages (preoperative, 2 weeks, 3 months, 6 months, and 12 months). A stage specific segmentation model delineated related structures, and an automated pipeline extracted quantitative, composite, qualitative, and dynamic features. Binary logistic regression models, constructed with and without dynamic parameters, assessed their incremental predictive value for best-corrected visual acuity (BCVA). A multimodal DL model combining clinical variables, OCT-derived features, and raw OCT images was developed and benchmarked against regression models. Results: The segmentation model achieved high accuracy across all timepoints (mean Dice > 0.89). Univariate and multivariate analyses identified base diameter, ellipsoid zone integrity, and macular hole area as significant BCVA predictors (P < 0.05). Incorporating dynamic recovery rates consistently improved logistic regression AUC, especially at the 3-month follow-up. The multimodal DL model outperformed logistic regression, yielding higher AUCs and overall accuracy at each stage. The difference is as high as 0.12, demonstrating the complementary value of raw image volume and dynamic parameters. Conclusions: Integrating dynamic parameters into the multimodal DL model significantly enhances the accuracy of predictions. This fully automated process therefore represents a promising clinical decision support tool for personalized postoperative management in macular hole surgery.

eess.IV↗

CLAPS: A CLIP-Unified Auto-Prompt Segmentation for Multi-Modal Retinal Imaging

Recent advancements in foundation models, such as the Segment Anything Model (SAM), have significantly impacted medical image segmentation, especially in retinal imaging, where precise segmentation is vital for diagnosis. Despite this progress, current methods face critical challenges: 1) modality ambiguity in textual disease descriptions, 2) a continued reliance on manual prompting for SAM-based workflows, and 3) a lack of a unified framework, with most methods being modality- and task-specific. To overcome these hurdles, we propose CLIP-unified Auto-Prompt Segmentation (\CLAPS), a novel method for unified segmentation across diverse tasks and modalities in retinal imaging. Our approach begins by pre-training a CLIP-based image encoder on a large, multi-modal retinal dataset to handle data scarcity and distribution imbalance. We then leverage GroundingDINO to automatically generate spatial bounding box prompts by detecting local lesions. To unify tasks and resolve ambiguity, we use text prompts enhanced with a unique "modality signature" for each imaging modality. Ultimately, these automated textual and spatial prompts guide SAM to execute precise segmentation, creating a fully automated and unified pipeline. Extensive experiments on 12 diverse datasets across 11 critical segmentation categories show that CLAPS achieves performance on par with specialized expert models while surpassing existing benchmarks across most metrics, demonstrating its broad generalizability as a foundation model.

cs.CV↗

UOPSL: Unpaired OCT Predilection Sites Learning for Fundus Image Diagnosis Augmentation

Significant advancements in AI-driven multimodal medical image diagnosis have led to substantial improvements in ophthalmic disease identification in recent years. However, acquiring paired multimodal ophthalmic images remains prohibitively expensive. While fundus photography is simple and cost-effective, the limited availability of OCT data and inherent modality imbalance hinder further progress. Conventional approaches that rely solely on fundus or textual features often fail to capture fine-grained spatial information, as each imaging modality provides distinct cues about lesion predilection sites. In this study, we propose a novel unpaired multimodal framework \UOPSL that utilizes extensive OCT-derived spatial priors to dynamically identify predilection sites, enhancing fundus image-based disease recognition. Our approach bridges unpaired fundus and OCTs via extended disease text descriptions. Initially, we employ contrastive learning on a large corpus of unpaired OCT and fundus images while simultaneously learning the predilection sites matrix in the OCT latent space. Through extensive optimization, this matrix captures lesion localization patterns within the OCT feature space. During the fine-tuning or inference phase of the downstream classification task based solely on fundus images, where paired OCT data is unavailable, we eliminate OCT input and utilize the predilection sites matrix to assist in fundus image classification learning. Extensive experiments conducted on 9 diverse datasets across 28 critical categories demonstrate that our framework outperforms existing benchmarks.

cs.CV↗

Semi-stable and splitting models for unitary Shimura varieties over ramified places. I

We consider Shimura varieties associated to a unitary group of signature $(n-s,s)$ where $n$ is even. For these varieties, we construct smooth $p$-adic integral models for $s=1$ and regular $p$-adic integral models for $s=2$ and $s=3$ over odd primes $p$ which ramify in the imaginary quadratic field with level subgroup at $p$ given by the stabilizer of a $π$-modular lattice in the hermitian space. Our construction, which has an explicit moduli-theoretic description, is given by an explicit resolution of a corresponding local model.

math.NT↗

Semi-stable and splitting models for unitary Shimura varieties over ramified places. II

We consider Shimura varieties associated to a unitary group of signature $(n-1, 1)$. For these varieties, we construct $p$-adic integral models over odd primes $p$ which ramify in the imaginary quadratic field with level subgroup at $p$ given by the stabilizer of a vertex lattice in the hermitian space. Our models are given by a variation of the construction of the splitting models of Pappas-Rapoport and they have a simple moduli theoretic description. By an explicit calculation, we show that these splitting models are normal, flat, Cohen-Macaulay and with reduced special fiber. In fact, they have relatively simple singularities: we show that a single blow-up along a smooth codimension one subvariety of the special fiber produces a semi-stable model. This also implies the existence of semi-stable models of the corresponding Shimura varieties.

math.NT↗

WOMD-Reasoning: A Large-Scale Dataset for Interaction Reasoning in Driving

Language models uncover unprecedented abilities in analyzing driving scenarios, owing to their limitless knowledge accumulated from text-based pre-training. Naturally, they should particularly excel in analyzing rule-based interactions, such as those triggered by traffic laws, which are well documented in texts. However, such interaction analysis remains underexplored due to the lack of dedicated language datasets that address it. Therefore, we propose Waymo Open Motion Dataset-Reasoning (WOMD-Reasoning), a comprehensive large-scale Q&As dataset built on WOMD focusing on describing and reasoning traffic rule-induced interactions in driving scenarios. WOMD-Reasoning also presents by far the largest multi-modal Q&A dataset, with 3 million Q&As on real-world driving scenarios, covering a wide range of driving topics from map descriptions and motion status descriptions to narratives and analyses of agents' interactions, behaviors, and intentions. To showcase the applications of WOMD-Reasoning, we design Motion-LLaVA, a motion-language model fine-tuned on WOMD-Reasoning. Quantitative and qualitative evaluations are performed on WOMD-Reasoning dataset as well as the outputs of Motion-LLaVA, supporting the data quality and wide applications of WOMD-Reasoning, in interaction predictions, traffic rule compliance plannings, etc. The dataset and its vision modal extension are available on https://waymo.com/open/download/. The codes & prompts to build it are available on https://github.com/yhli123/WOMD-Reasoning.

cs.RO↗