Searcharxiv⌕ Search

arXiv subjects

Zihan Li

Publications and source records attributed to Zihan Li.

At least 19 recordsLinked to original sources

REGER: Modeling inter-item Relations with Semantic IDs for Generative Recommendation

Personalized recommendation depends on learning semantic and collaborative relations among items from online content and user interactions. Generative recommendation (GR) offers a promising paradigm by autoregressively generating discrete semantic IDs (SIDs) of items. However, randomly initialized SID token embeddings and GR models trained from scratch do not directly inherit the semantic relations inherent in items, while inter-item collaborative relations are learned largely implicitly through token-level attention and next-token supervision. To address this limitation, we propose REGER, a unified framework that explicitly incorporates inter-item relations into semantic pretraining, history encoding, and SID generation. We pretrain the GR model through generative semantic retrieval, enabling it to inherit inter-item semantic similarity by learning to generate semantic neighbors' SIDs. We combine global attention with local attention aligned with SID boundaries to capture historical context and directed transitions between adjacent items. We formalize item-level preference distillation on the SID prefix tree, deriving an exact factorization of item preference distributions into prefix conditionals and a weighted decomposition of item-level KL divergence into conditional KL terms. Guided by this decomposition, we renormalize the generator's probabilities over legal next tokens and match the resulting distributions to the teacher's conditional preferences along the ground-truth SID path, jointly optimizing the distillation objective with cross-entropy. Extensive experiments on real-world datasets validate the effectiveness of the proposed preference distillation and demonstrate that REGER consistently outperforms existing GR baselines while retaining efficient inference.

cs.IR↗

SemLite-GS: Semantic-Selective Parallel Gaussian Splatting for Lightweight Building Reconstruction from UAV Oblique Imagery

Lightweight 3D building models support urban spatial analysis, but their production often requires dense scene reconstruction, building extraction, and mesh simplification. Although 3D Gaussian Splatting (3DGS) is more efficient than conventional multi-view stereo, existing large-scale methods lack semantic selection and parallel instance-level training. We present SemLite-GS, a building-centric framework that reconstructs lightweight building meshes from large-scale UAV oblique imagery. First, an orthographic reference map establishes globally consistent building identities, which are propagated to oblique views to form camera-instance pairs and regions of interest. Second, semantic-selective parallel PGSR training jointly processes multiple pairs within relevant image regions while updating only target Gaussians; contextual Gaussians remain fixed to preserve visibility. Third, architectural planes extracted from rendered depth, opacity, and normal maps guide signed distance field optimization, producing meshes with adaptive polygon allocation and well-defined structural boundaries. Experiments on multiple large-scale UAV datasets demonstrate that our SemLite-GS substantially reduces building reconstruction time compared with representative full-scene 3DGS methods. With comparable polygon budgets, SemLite-GS achieves superior preservation of dominant planar structures and sharp architectural transitions compared with conventional dense reconstruction followed by mesh decimation. The source code is available at https://github.com/IRIS-LAB-whu/SemLite-GS.

cs.CV↗

Larch: Learned Query Optimization for Semantic Predicates

With the advent of Large Language Models (LLMs), many database systems introduced semantic operators that enabled analytical queries over unstructured data (e.g. text, images, videos). Semantic operators typically incur high inference costs and latencies making semantic (AI) SQL queries challenging to apply on large scale datasets. At the same time, their semantic nature leads database engines to treat them as black boxes, making AISQL queries difficult to optimize. In this paper, we introduce Larch, a framework for optimizing the execution of semantic filters in AI SQL queries. Larch was inspired by two key observations: i) the high latency of semantic operators leaves significant room for computationally-heavy runtime optimization techniques, ii) unstructured data are typically accompanied by semantic information in the form of embeddings allowing for efficient semantic comparisons between AI_FILTER prompts and data values. Based on these two key observations, we present two Larch variants: Larch-A2C and Larch-Sel. Larch-A2C encodes arbitrary semantic filters expression tree using an embedding-augmented Gated Graph Neural Network and formulates the filter evaluation order as a Markov decision process. In contrast, Larch-Sel leverages a supervised learning model to predict filter selectivities, subsequently applying dynamic programming to find a near-optimal evaluation order for each input row. Evaluated across diverse real-world datasets and comprehensive synthetic workloads, both Larch variants always outperform existing semantic filter optimization techniques in terms of token usage on every dataset and workload. Our results demonstrate that Larch is robust across diverse workloads across text and image modalities, 6%--30% lower total token costs and 3x--19x lower token costs overhead than Palimpzest and Quest.

cs.DB↗

Generalizable fetal brain biometry via adaptive artificial intelligence

Fetal brain biometry provides a quantitative basis for assessing prenatal growth and identifying congenital brain abnormalities. However, comprehensive assessment remains limited by labor-intensive manual measurements and automated models that lose accuracy across imaging sequences and institutions. We introduce BioTTA, an anatomically guided test-time adaptation framework that measures 11 fetal brain biometrics from reconstructed three-dimensional magnetic resonance images. BioTTA adapts pretrained models to individual unlabeled volumes using geometric consistency and atlas priors, without access to source images. In a retrospective study across five centers in China and Switzerland, BioTTA improved measurement accuracy across three out-of-distribution datasets and two network architectures. It reduced mean absolute measurement errors by 46-82% relative to the corresponding models without adaptation, achieving errors similar to those of supervised fine-tuning. Its measurements were mapped onto normative growth trajectories derived from expert annotations, enabling gestational-age-specific assessment through a web-based reporting platform. Across four disease cohorts, the platform achieved 82.1-95.0% accuracy in binary detection of ventriculomegaly or cerebellar abnormalities, with performance similar to that of a senior radiologist. BioTTA provides a practical foundation for standardized fetal brain biometry across heterogeneous imaging settings, supporting scalable prenatal growth assessment and quantitative evaluation of brain abnormalities.

cs.CV↗

RoadState: A Domain-Prior-Enhanced Vision-Language System for Industrial-Grade Lane-Level Map Updating

Lane-level map updating requires temporally consistent, physically valid, and auditable road-scene semantics despite occlusions, ambiguous lane markings, temporary obstacles, and viewpoint changes. We present RoadState, a production-deployed vision-language system that transforms sequential road-scene observations into structured lane-level semantics for Baidu Maps' map-updating pipeline. Its core model, RoadState-VLM, formulates road-scene understanding as autoregressive state estimation, predicting the current structured road state from the previous observation and symbolic state together with the current observation. We introduce Geometric Chain-of-Thought (Geo-CoT), which uses a directed acyclic graph of semantic dependencies as a domain prior to improve semantic consistency and reasoning fidelity across interdependent road-state dimensions. Extensive experiments on a real-world dataset show that RoadState-VLM achieves 94.33% mean accuracy across six road-state dimensions, outperforming the strongest fine-tuned VLM baseline. Ablation studies further demonstrate the complementary effectiveness of Geo-CoT and temporal transition context. A cascaded routing pipeline directs routine cases to lightweight models while reserving complex cases for RoadState-VLM. Since its deployment in March 2025, RoadState has supported lane-level map updating across more than 360 cities and reduced overall production costs by 30%. Together with human-in-the-loop auditing, the system maintains robust performance under stress-test conditions. These results demonstrate how domain-prior-enhanced reasoning, structured state interfaces, selective routing, and human quality control can bridge the gap between VLM-based road-scene understanding and large-scale lane-level map updating.

cs.CV↗

EXACT: an explainable anomaly-aware vision foundation model for analysis of 3D chest CT

Chest computed tomography (CT) is central to the detection and management of thoracic disease, yet the growing scale and complexity of volumetric imaging increasingly exceed what can be addressed by scan-level prediction alone. Clinically useful AI for CT must not only recognize disease across the whole volume, but also localize abnormalities and provide interpretable visual evidence. Existing vision-language foundation models typically compress scans and reports into global image-text representations, limiting their ability to preserve spatial evidence and support clinically meaningful interpretation. Here we developed EXACT, an explainable anomaly-aware foundation model for three-dimensional chest CT that learns spatially resolved representations from paired clinical scans and radiology reports. EXACT was pre-trained on 25,692 CT-reports pairs using anatomy-aware weak supervision, jointly learning organ segmentation and multi-instance anomaly localization without manual voxel-level annotations. The resulting organ-specific anomaly-aware maps assign each voxel a disease-specific anomaly score confined to its corresponding anatomy, jointly encoding lesion extent and organ-level context. In retrospective multinational and multi-center evaluations, EXACT showed broad and consistent improvements across clinically relevant CT tasks, spanning multi-disease diagnosis, zero-shot anomaly localization, downstream adaptation, and visually grounded report generation, outperforming existing three-dimensional medical foundation models. By transforming routine clinical CT scans and free-text reports into explainable voxel-level representations, EXACT establishes a scalable paradigm for trustworthy volumetric medical AI.

cs.CV↗

An integrated photonic millimeter-wave receiver with sub-ambient noise temperature

Growing demands for timing precision in positioning systems, and resolution in atmospheric sensing and automotive radar, have extended receiver frontends into the millimetre-wave and THz regimes. At these frequencies, the noise performance of field-effect transistors used in low-noise amplifiers degrades rapidly due to parasitic effects, limited carrier mobility, hot electrons, and shot noise. Parametric transducers that couple electromagnetic signals to optical fields can offer quantum-limited sensitivity at room temperature. Electro-optic materials can enable receivers that transduce radiofrequency signals into optical phase shifts. Yet to date, room-temperature receivers using electro-optical detection have not attained a noise figure comparable to the electronic amplifiers. Here we demonstrate a room-temperature cavity electro-optic-based millimetre-wave receiver on a lithium tantalate photonic integrated circuit, achieving an on-chip photon-number transduction efficiency of 10%. The equivalent input-referred receiver-added noise temperatures reach $110\,\mathrm{K}$ on-chip and $202^{+35}_{-33}\,\mathrm{K}$ for the full system at 60.5 GHz, with noise figure matching state-of-the-art V-band electronic low-noise amplifiers. We resolve the thermal noise in cavity electro-optic transducers, showing that the system is limited by thermal photon occupation ($\sim$100) in the millimetre-wave cavity and optical vacuum noise, with no measurable excess noise. Our work demonstrates that integrated photonics can reach state-of-the-art electronic receiver noise performance while offering high-radiofrequency-power survivability and compatibility with remote optical readout. This establishes cavity electro-optics as a low-noise, electromagnetic interference-resilient, chip-scale receiver frontend for millimetre-wave applications, enabling scalable analogue processing in the optical domain.

physics.optics↗

Direct Action-Manifold Modeling for Efficient and Robust Bimanual Manipulation

Diffusion and flow matching visuomotor policies require iterative denoising or numerical integration for multimodal action generation, introducing latency and execution mismatch. We instead consider a different formulation: conditioned on vision and proprioception, demonstrated bimanual action chunks concentrate in a structured observation-dependent region of joint-action space, making the action target more direct to model than the generative process. Based on this view, we propose the Direct Manifold Action Transformer (DMAT), which reformulates action generation as the direct modeling of clean, observation-conditioned joint-space action chunks from stochastic interpolants, thereby enabling single-step inference without relying on iterative generative inference or auxiliary trajectory-straightening procedures. Within DMAT, local 3D geometry anchors predictions to spatial structure, while State-Consistent Spectral-Time Refinement (STR) reconciles predicted chunks with proprioception and suppresses high-frequency residuals. Across 567 standard and out-of-distribution (OOD) trials per method, DMAT outperformed Improved 3D Diffusion Policy (IDP3) and DiTFlow3D in aggregate success rate, achieving relative improvements of 129.1% and 95.7%, respectively, while generating action chunks 2.06x and 1.34x faster and remaining robust under repeated 50 N interventions. These results pave the way for a novel generative paradigm for efficient and robust bimanual manipulation.

cs.RO↗

safeL4: Rethinking Safe-Language-Based Single Address Space Microkernels

Microkernels provide good features such as high security, safety, and flexibility, but their reliance on hardware-based isolation leads to heavy performance overhead, limiting their widespread adoption. Although constructing single-address-space microkernels based on safe language-based isolation is a promising approach to addressing the performance issue, this approach still faces challenges in maintaining the principle of minimality and enforcing capability-based, fine-grained access control over system resources. We propose safeL4, an approach to constructing microkernels based on safe languages. It guarantees minimality through low-level abstractions for unsafe operations, and implements a capability system based on the safe language's object capabilities to achieve fine-grained access control over all operations on system resources. We implement a prototype system in Rust, and our evaluation results show that the primitives provided by safeL4 are, on average, 1.9$\times$ faster than those of seL4.

cs.OS↗

Who Verifies the Agent's Success Claim? Self-Report Leakage in GUI-Agent Completion Judging

In automation pipelines, a GUI agent's confident success report can let downstream stages proceed even when the recorded execution trace does not show that the agent produced the requested outcome. We call this case a dangerous misreport: an unhedged success claim that the trace does not support as causal accomplishment during the run. We study report--trace auditing, which checks an agent's completion claim against an independently recorded trace. A controlled aware/blind intervention reveals self-report leakage: the agent's own self-report changes how completion is judged. The same judge reads the same task and trace; only that self-report is included or withheld. Withholding it improves trace-only dangerous-misreport recall by 43.4 percentage points with Qwen3.5-35B-A3B and 13.3 percentage points with DeepSeek-284B-A13B, with the same trend on held-out Batch 2. This motivates report isolation. The report records what the agent claims, but it must not influence the trace-state judgment. We implement this separation with a structured auditor that separates claim interpretation from claim-blind trace analysis and applies deterministic rules to extracted evidence. Because evidence extraction is separated from the policy that consumes it, the same audit record can be replayed under alternative policies without new model calls. The auditor achieves F1 0.75--0.78 on the 465-record audit-target set and 0.77--0.81 on held-out Batch 2, outperforming claim-aware and prompt-blind controls in recall and F1. Under audit-stage gating, where flag and review both block automatic pass, it blocks 74--90% of dangerous misreports at 45--56% inspection load, or 16--19% of the full in-scope stream after coarse status triage.

cs.SE↗

Lattices with Large Kissing Number from Local Density

Let $κ_p(n)$ denote the maximum kissing number of a lattice $\mathcal{L} \subset \mathbb{R}^n$. We prove that $κ_p(n) \geq 2^{Ω(\sqrt{n})}$ for all $p\ge 1$ and that $κ_p(n) \geq Ω(α_p^n)$ for all $p > 2$, where $α_p > 1$ is a constant depending only on $p$. In particular, our former result improves on the previously best known lower bound of $κ_2(n) \geq 2^{Θ(\log^2 n)}$ for the $\ell_2$ kissing number, which has stood for over $60$ years. (In concurrent and independent work, Laarhoven and Kominers [arXiv, 2026] showed a similar lower bound on $κ_2(n)$ using similar techniques.) In fact, our lower bounds follow from efficient (randomized) constructions of lattices with large kissing number. Our results use a general construction of lattices with large kissing number. Specifically, we show a simple equivalence between such lattices and a new object that we call an exact locally dense lattice. Closely related objects appear throughout the literature on computational lattice problems. Our constructions of lattices with large kissing number then follow as corollaries by showing that constructions due to Bennett and Peikert [RANDOM, 2023] (using Craig lattices) and Aggarwal and Stephens-Davidowitz [STOC, 2018] (using sparsification of $\mathbb{Z}^n$) in fact give exact locally dense lattices, as well as a new construction of exact locally dense lattices built from rotations of the $m$-dimensional checkerboard lattice $D_m$.

math.MG↗

PageWeaver: KV-Guided Query Unions for Sparse Attention

Dynamic sparse attention limits the KV pages selected by each query, but a small support does not necessarily yield efficient GPU work. Query unions share page loads and populate Tensor Core tiles; their cost depends on which queries are grouped together. We present PageWeaver, an execution design that uses selected-page affinity to assemble query groups while preserving each query's original support and complete output ownership. A bounded GPU search produces query IDs, and an ID-aware two-CTA kernel consumes them without materializing reordered Q tensors or cross-page partial outputs. A direct KV-page union implementation provides a complementary design study of nonlocal reuse and reduction cost. With FP8 KV throughout, the H200 Union8 implementation achieves a 1.70x geometric-mean complete-call speedup over the measured FlashInfer path on six captures. Online regrouping further lowers latency by 3.26-7.66% on five selected 64K-context captures. Whole-model prefill throughput is 7.88-14.36% above the tested native path; the incremental regrouping benefit is smaller, with observed median gains of 0.47-0.73% at 32K/64K and regressions at 8K. A B300 comparison identifies cases where preparation cost and a stronger native kernel remove the advantage. These results separate execution-group reuse from the complete cost of exploiting it online.

cs.DC↗

Multimodal Deep Survival Analysis for Sinkhole Susceptibility

Sinkholes are a widespread geohazard in karst terrain. In Florida, soluble carbonate bedrock, shallow groundwater, and intense rainfall combine to make subsidence both common and spatially heterogeneous. Predicting where and when sinkholes will occur is difficult for two reasons. First, locations without reported sinkholes cannot be directly labeled or sampled as true negative locations. Second, the potential factors governing sinkhole risk span heterogeneous data modalities and therefore require careful integration within a unified modeling framework. We address both problems with our proposed model, a multimodal Cox proportional hazards framework for sinkhole susceptibility. Our contributions are threefold. First, we extend the proportional-hazards formulation to heterogeneous multimodal input through modality-specific encoders and a cross-modal fusion layer. Second, we treat unreported locations as right-censored rather than negative, avoiding hard-negative labeling and yielding continuous, time-aware susceptibility from the predicted survival function. Third, a statewide Florida case study with spatially blocked validation and ablation studies quantifies the benefit of multimodal integration. A Florida case study demonstrates that the proposed method effectively ranks sinkhole risk and produces a statewide susceptibility map that captures spatial variations in sinkhole occurrence.

cs.LG↗

templar: agentic induction and evolution of standardized radiology reporting templates from large-scale clinical corpora

Structured radiology reporting mitigates the heterogeneity of free-text reports, yet its benefits depend on high-quality reporting templates. In practice, such templates are conventionally built through labor-intensive expert consensus and therefore vary across institutions and lag behind evolving clinical practice. Large language models (LLMs) enable automated template induction, but existing approaches remain limited: single-LLM induction is constrained by context length, and the corpus-scale method ASTAR produces a static, closed-corpus template without external grounding or downstream adaptation. To address these limitations, we propose TEMPLAR, a TEMPLate-centric Agentic framework for inducing and evolving standardized Radiology reporting templates from large-scale clinical corpora. TEMPLAR treats the template as a persistent central state maintained alongside two provenance-aware knowledge graphs, namely an anatomical graph that constrains template construction and a diagnostic graph that supports finding-to-diagnosis reasoning. Three agents operate on this state. The Induction Agent derives canonical clinical slots from anatomy-constrained Span-Triple atoms via dual-view similarity clustering; the Evolution Agent then assembles these slots into a hierarchical template and revises it under consistency constraints, external clinical evidence, and downstream structuring feedback; and the Clinical Agent applies the evolved template to report structuring, reconstruction, and diagnostic reasoning. Across four datasets, TEMPLAR outperforms ASTAR, three medical LLMs, and six general-purpose LLMs in coverage, information fidelity, and diagnostic fidelity, while achieving the highest or tied-highest LLM-rated template quality. Its fidelity advantages over ASTAR persist under cross-dataset transfer, and cumulative ablations support complementary contributions of its key components.

cs.MA↗

Hierarchical Continuous Diffusion Language Models

Discrete diffusion language models offer a compelling alternative to autoregressive generation for tasks demanding bidirectional reasoning and global constraint satisfaction. Yet they share a structural bottleneck: when decoding in parallel, each token is sampled independently from its marginal, severing the statistical dependencies among the tokens decoded together. Continuous diffusion language models avoid this by denoising a shared continuous state, but their denoiser sees only that state, so nothing ties it to a valid token configuration until it is finally decoded. To address this, we propose Hierarchical Continuous Diffusion Language Models (HC-DLM), which couple discrete token generation with a continuous latent trajectory in a single, principled denoising process, whose training objective is derived from a variational bound on the token likelihood. In contrast to recent methods that attach continuous context to a self-contained discrete chain, HC-DLM makes the latent the only persistent generative state: tokens are read out from it at every step and feed back as a scaffold for the next latent update. On structured reasoning (Sudoku), mathematical planning (Countdown) and language modeling (LM1B), HC-DLM improves over discrete and continuous diffusion baselines at matched model size, in puzzle accuracy on Sudoku and Countdown and in generative perplexity on LM1B. Project page: https://hc-dlm.github.io/.

cs.CL↗

RetroHolmes: When Semantic Plausibility Fails Retrospective Physical Process Reasoning

Vision-Language Models (VLMs) are widely used for visual understanding, yet current evaluation protocols fail to assess whether these capabilities are grounded in physical reasoning. To address this gap, we introduce Retrospective Physical Process Reasoning, a new evaluation paradigm to reason backward from outcomes under explicit physical constraints. Building on the paradigm, we present RetroHolmes, the first real-world benchmark for Retrospective Physical Process Reasoning, comprising object-centric image pairs annotated with reachability labels and causal step sequences across diverse physical transitions. Using RetroHolmes, we analyze VLMs and uncover systematic failure modes, including judgment bias in reachability assessment and belief dominance over physical evidence, mirroring sycophancy behavior observed in large language models. Our quantitative analyses link these failures to reliance on linguistic priors and attention concentrated on visually invariant regions, suggesting limited physical simulation of the intermediate states connecting visual endpoints. To address these limitations, we propose Simulate-and-Verify, an analysis-by-synthesis framework that grounds reachability judgment and step reconstruction in visual simulation. Experiments show that Simulate-and-Verify improves judgment accuracy by 21.67 percentage points and reduces belief dominance by 10.39 percentage points compared with GPT-5.5, demonstrating the effectiveness of visual simulation in grounding physical reasoning.

cs.MM↗

NeuSOGA3D: A Neuro-Symbolic Framework for Explainable 3D Geometric Reconstruction

Three-dimensional reconstruction from unorganized point clouds remains a challenging problem in computer vision, geometric modeling, and computer-aided design. While neural implicit methods achieve impressive reconstruction accuracy, geometry is typically encoded in latent representations that limit interpretability and reuse within engineering workflows. We present NeuSOGA3D (Neuro-Symbolic Observation-Guided Geometric Abstraction in 3D), a hybrid framework that combines learned perceptual priors inherited from NeuSOGA with explicit symbolic geometric reasoning. The method projects point clouds onto principal orthographic planes, constructs symbolic implicit spline representations from the resulting observations, and fuses them through shape-preserving constructive solid geometry operations to generate a coarse visual hull. Additional geometric detail is recovered through cross-sectional decomposition and volumetric reconstruction using Partial Shape-Preserving Splines. Unlike conventional neural implicit approaches, NeuSOGA3D progressively transforms observations into explicit symbolic entities, including control polygons, implicit spline fields, cross-sections, and volumetric lofts. Experiments on all forty categories of the ModelNet40 benchmark demonstrate the ability of the framework to recover structurally meaningful and CAD-compatible geometric representations from diverse point-cloud observations. The results highlight the potential of combining learned perception with symbolic geometric reasoning for explainable geometric intelligence.

cs.AI↗

Galactic constraints on spinning primordial black hole dark matter from interstellar dust heating

Primordial black holes (PBHs) are compelling dark matter candidates. PBHs with masses between $10^{15}$ and $10^{18}\,\mathrm{g}$ can heat interstellar dust via Hawking radiation. Previous studies of this dust heating mechanism mostly neglected PBH spin and adopted a single dark matter halo profile. In this work, we incorporate PBH spin, which substantially enhances the emitted radiation flux, and systematically investigate the dependence of constraints on the dark matter density distribution by considering five different halo models. We compute the complete photon spectra, including both primary and secondary emissions. Our results show that, for a fixed profile and mass function, larger spin parameters yield stronger constraints on the PBH fraction $f_{\mathrm{PBH}}$. Among the halo models, the Isothermal profile gives the most stringent limits, followed by Einasto, then NFW and Moore, while the Burkert profile yields the weakest constraints. For silicate grains, which cool less efficiently than graphite, the upper limits reach $\mathcal{O}(10^{-4})$ for high spin cases. We consider both monochromatic and lognormal mass functions, and find consistent trends between them. For the lognormal case, larger values of the width $σ$ lead to a broader mass range being excluded, in particular ruling out massive PBHs as the sole dark matter component. Our bounds are generally weaker than other existing limits, but they provide a complementary and independent constraint.

astro-ph.CO↗