Search arXiv⌕ Search

arXiv subjects

Tianyi Yang

Publications and source records attributed to Tianyi Yang.

At least 19 recordsLinked to original sources
arXiv2026

Perceptual Distortion for Overfitted Video Coding: Sliced Matching in Space, Error Alignment in Time

Recent works in neural video coding have increasingly been trading off perceptual quality against reconstruction fidelity at a fixed bitrate. Wasserstein Distortion (WD) offers a principled approach to balancing these objectives by relaxing spatial correspondences in feature matching, enabling lightweight overfitted image codecs to compete with computationally intensive generative decoders. Extending vanilla WD to video, however, presents two challenges: its independent Gaussian approximation limits the expressiveness of feature distribution matching, while unconstrained texture resampling across frames introduces temporal inconsistencies. We address these challenges by distinguishing between the correspondences that can be relaxed and those that must be preserved. Spatially, we introduce a local sliced Wasserstein distortion (SWD) that matches empirical feature distributions through one-dimensional optimal transport, capturing cross-channel dependencies without enforcing exact spatial alignment. Temporally, we employ a novel flow residual that constrains coding errors along motion trajectories, preserving temporal consistency without imposing additional reconstruction fidelity constraints. Evaluated on UVG and HEVC Class B with the overfitted codec COOL-CHIC, our framework achieves BD-rate reductions of 9.2% in DISTS and 3.0% in LPIPS relative to frame-wise WD, averaged over test sequences, while lowering tOF by 10.2% at matched rate.

cs.CV↗
arXiv2026

Priority-Aware Scheduler for Serverless Functions Provision and Load Balancing

Serverless computing has emerged as a pivotal cloud deployment paradigm due to its efficient utilization of on-demand resources. However, the persistent cold start problem continues to hinder system performance. Current strategies often employ oversimplified preloading/unloading mechanisms that inadequately exploit invocation patterns, failing to optimize the critical trade-off between startup latency and resource efficiency. Furthermore, existing approaches disregard practical constraints of memory-limited multi-worker environments, exacerbating function eviction and load imbalance issues. To address these limitations, we present SPES-Ext, a rule-based priority-aware scheduler designed for memory-constrained distributed worker nodes. Our methodology stems from two key observations: (1) serverless systems exhibit concentrated invocation patterns enabling behavioral predictability, and (2) these patterns permit function categorization for granular instance management. This insight posits that characteristic-driven prioritization enables optimal scheduling rules, leading to our system design, which integrates three coordinated components: predictive function provisioning, eviction optimization, and dynamic load balancing. Experimental validation demonstrates the superiority of SPES-Ext over state-of-the-art solutions: 80% reduction in $75^{th}$-percentile cold starts, 83.92% decrease in memory waste, and 52.49% improvement in worker load variance.

cs.DC↗
arXiv2026

PosterHarness: Turning Scientific Poster Generation into an Auditable Instruction-Following Benchmark

Text-rich image models can now design poster-scale layouts, but we lack ways to measure whether they honor scientific communication contracts: legible labels, prescribed aspect ratios, and -- above all -- abstaining from fabricated scientific figures. We present POSTERHARNESS, an auditable harness reframing poster generation as measurable instruction-following tasks, with a pilot benchmark and failure taxonomy. POSTERHARNESS uses a placeholder-first contract to separate two jobs models otherwise conflate. The model performs visual-summary design: typography, reading path, color, and background -- but never draws data-bearing figures. Every figure region must be an empty labeled placeholder; a deterministic compositor inserts real source-paper figures at detected coordinates. This makes properties measurable: placeholder count and ID accuracy, blankness, aspect-ratio compliance, abstention from synthesized graphics, public-text hygiene, and source-figure provenance -- with failures logged as explicit rejections, not hidden in plausible-looking output. We instantiate the harness on 12 papers (6 HEP, 6 AI/ML-adjacent) and report three findings. (i) A counterfactual probe shows the placeholder contract drives VLM-counted synthesized figures from 34 to 0 across three papers. (ii) A failure taxonomy identifies blocking contracts: placeholder geometry, placeholder QA, template critic, and public text. (iii) Comparison with Paper2Poster shows a trade-off: PosterHarness yields higher-resolution artifacts, lower white-canvas fraction, and stronger VLM visual preference; the deterministic baseline retains slightly more PosterQuiz-style information and runs faster. We report this as regime characterization, not a superiority claim. All artifacts, prompts, manifests, and audit scripts are released as a reusable evaluation component.

cs.CV↗
arXiv2026

Structure-Aware Spatial Kernels for Gaussian Process Distributed Port-Hamiltonian Systems

Gaussian process distributed port-Hamiltonian systems (GP-dPHS) enable data-driven modeling of distributed dynamical systems while preserving their physical structure. However, standard stationary kernels can be inefficient for spatially heterogeneous systems because they neither exploit known input invariances nor reflect structured spatial dependencies. We address this limitation with a structure-aware kernel design that removes irrelevant state coordinates and decouples state and spatial dependencies through a tensor-product covariance. This modular construction allows qualitative knowledge of spatial heterogeneity to be incorporated by selecting an appropriate spatial kernel without requiring an analytical description of the underlying profile. Numerical experiments on a nonlinear string with periodic, smoothly varying, and semi-piecewise stiffness profiles demonstrate that structural decoupling consistently improves reconstruction accuracy, while profile-matched spatial kernels yield substantially larger gains in force-field and trajectory reconstruction.

cs.LG↗
arXiv2026

OpenDepth: Repurposing Image Editing Priors for Robust Open-World Monocular Depth Estimation

Recently, the integration of generative priors has revolutionized Monocular Depth Estimation (MDE), achieving state-of-the-art (SOTA) performance with remarkably high data efficiency. However, although methods based on Text-to-Image (T2I) models excel in general scenes, their efficacy degrades sharply when confronted with open-world environmental perturbations, such as illumination variations and adverse weather. Since T2I priors are optimized for semantic synthesis rather than preserving pixel-level structure, they struggle to decouple geometry from appearance under severe visual shifts, easily triggering structural hallucinations. Therefore, our core insight is to model open-world depth estimation as consistent geometric inference across multi-style images, fully leveraging the powerful structure-preserving editing priors of Image-to-Image (I2I) models. Based on this, we propose OpenDepth, a robust open-world MDE framework that introduces "Intra-Domain Awareness" -- the capability to map diverse stylistic inputs sharing identical geometry into a unified latent space. To achieve this, we first align shallow features within the Diffusion Transformer (DiT) to anchor degraded observations to unified geometric representations. Additionally, we introduce a joint RGB latent reconstruction task, requiring the model to recover clean RGB information from degraded inputs. This multi-stage strategy effectively decouples environmental style interference, compelling the model to focus strictly on underlying geometric recovery. Extensive evaluations show that OpenDepth sets new SOTA records for robust zero-shot depth estimation across standard and extreme weather benchmarks, demonstrating exceptional structural integrity and adaptability.

cs.AI↗
arXiv2026

Drive3R: Block-wise Streaming Geometric Foundation Models for Autonomous Driving

High-fidelity closed-loop simulation for autonomous driving increasingly relies on dense multi-view 3D reconstruction. While recent feed-forward geometric foundation models have revolutionized per-scene optimization, applying them to continuous, long-horizon driving sequences is computationally intractable due to GPU memory limits and the quadratic complexity of global attention. Existing scaling strategies are fundamentally limited, as causal streaming suffers from compounding scale drift and trajectory deviation, while rigid "chunk-then-align" heuristics fragment global context and produce trajectory discontinuities. To address these limitations, we introduce Drive3R, a block-wise streaming geometric foundation model. Its block-wise causal attention resolves the scalability bottleneck of fully bidirectional attention while avoiding the loss of local future context in purely causal streaming. Its confidence-aware re-anchoring further mitigates long-range drift by detecting unreliable streaming states, re-initializing the KV cache, and stitching refreshed segments through SE(3) alignment. Extensive experiments demonstrate that Drive3R establishes a new state-of-the-art on nuScenes, KITTI, and Waymo. Specifically, on nuScenes, it improves over the recent DVGT-2 baseline from 0.1806 to 0.1081 in AbsRel, from 0.6750 to 0.5344 in point-cloud Accuracy, and from 1.5812 to 0.9092 in ATE.

cs.CV↗
arXiv2026

A possible detection of gas rotation in galaxy groups via the kinetic Sunyaev-Zeldovich effect

We utilise the kinetic Sunyaev-Zeldovich effect (kSZ) to measure the rotation of ionised gas within galaxy groups defined in the SDSS-DR7 galaxy sample, via their dipolar imprint on the cosmic microwave background (CMB). We estimate the direction of the projected angular momentum for each of our selected fast-rotating groups by measuring the redshift dipole of satellite galaxies around their group centres. We find a clear redshift dipole in the stacked data for the selected SDSS groups. We then perform oriented stacking of the Planck CMB temperature map using the group centres and directions of angular momenta. We report a $2.3σ$ tentative detection of the dipolar rotational kSZ effect (rkSZ) within the virial radii of our fast-rotating groups from SDSS with an average mass of $10^{14}h^{-1} \rm M_{\odot}$ (corresponding to a null-hypothesis probability of 0.96%). We estimate the averaged rotational velocity of the sample to be $\sim 100-200 ~\rm km ~s^{-1}$, reaching a maximum value of $158.8\pm13.6 ~\rm km ~s^{-1}$. Our results are consistent within the errors with predictions based on the ELUCID constrained realisation simulation, with the predicted amplitude of the rkSZ signal being slightly lower near the centre. We also identify a systematic bias when estimating rotational velocities using the observed redshifts of galaxies, but find it to be subdominant for our analysis.

astro-ph.CO↗
arXiv2026

Reinterpreting ATLAS Dijet and Supersymmetry Searches in A Baryon-Number-Violating Heavy Scalar Framework

A reinterpretation of ATLAS searches targeting $b$-tagged dijet resonances and monotop-alike signatures at a center-of-mass energy of 13 TeV is presented using $pp$ collision data corresponding to integrated luminosities of 139 $\mathrm{fb}^{-1}$. These searches are used to constrain a baryon-number-violating new physics model that introduces two scalar mediators $X_{1}$ and $X_{2}$ to simultaneously address the baryon asymmetry of the universe and the dark matter relic abundance. By exploiting heavy-flavor signatures and allowing for jet-rich final states, this reinterpretation provides complementary sensitivity to the model relative to non-flavor-tagged inclusive dijet searches and conventional monotop reinterpretations. Assuming the mediators decay either into a pair of down-type quarks or into a top quark and a dark matter candidate, we translate the experimental upper limits into the model's parameter space. 95% confidence level exclusion limits are derived in the $λ_1$-$λ_2$ coupling plane for benchmark mediator masses of 1.5, 1.75, and 2.0 TeV. For all three benchmark masses, the $b$-tagged dijet reinterpretation excludes $λ_1 > 0.45$, while the bottom-squark reinterpretation excludes the region with $λ_1 > 0.3$ and $λ_2 > 0.6$, within the respective validated parameter domains.

hep-ph↗
arXiv2026

Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis

Muon collider research spans accelerator physics, detector instrumentation, and high-energy phenomenology, with relevant evidence scattered across a rapidly expanding and heterogeneous body of scientific literature. As high-energy physics (HEP) increasingly explores agent-assisted analysis workflows, efficiently locating, integrating, and verifying scientific evidence becomes an essential capability. While retrieval-augmented generation (RAG) offers a promising framework for scientific question answering, integrating agentic reasoning without compromising retrieval precision remains a key challenge. In this work, we present agentic hybrid RAG, an evidence-grounded RAG framework for muon collider research. The framework combines a hybrid retriever, integrating sparse lexical and dense semantic retrieval, with an agentic reasoning module for query decomposition, evidence expansion, and grounded answer generation. To enable systematic evaluation, we construct the first benchmark for retrieval-augmented scientific question answering in the muon collider domain, comprising a curated literature corpus together with dedicated retrieval and answer-generation benchmarks covering major detector and physics research topics. Extensive evaluation shows that hybrid retrieval provides the strongest retrieval backbone, while agentic reasoning is most effective for controlled evidence expansion and answer synthesis. Built on this principle, agentic hybrid RAG consistently outperforms representative retrieval and RAG baselines in retrieval effectiveness, answer quality, evidence coverage, and factual grounding. Together, the benchmark and framework provide a foundation for evidence-grounded scientific question answering and future HEP analysis agents operating over large-scale scientific literature. Code is available at \href{https://github.com/AItutorialjrb/RAG_muon_JINST}{this URL}.

hep-ex↗
arXiv2026

HelloWorld: Towards Practical Applications of Generative Driving World Models

Driving world models provide a promising route toward scalable counterfactual data generation and interactive simulation beyond recorded driving logs. Realizing this potential requires a system that can generalize across diverse scenes, respond faithfully to prescribed controls, generate coherent multi-sensor observations, and operate efficiently under repeated inference. We present \textbf{HelloWorld}, a 2B driving world model system designed around these requirements. HelloWorld progressively specializes broad visual and motion priors from heterogeneous video data into controllable driving generation using ego pose, HD maps, and 3D boxes. A block-causal generation interface, together with adaptation to self-generated context, aligns the model with sequential simulation. The system further supports synchronized seven-camera RGB generation and conditional LiDAR synthesis, and is distilled toward few-step inference for efficient deployment. Experiments evaluate visual quality, control fidelity, cross-view consistency, robustness under repeated generation, inference efficiency, and LiDAR synthesis. Together, HelloWorld provides a unified framework for scalable driving data generation and interactive simulation.

cs.CV↗
arXiv2026

Human-Inspired Neuro-Symbolic World Modeling and Logic Reasoning for Interpretable Safe UAV Landing Site Assessment

Reliable assessment of safe landing sites in unstructured environments is essential for deploying Unmanned Aerial Vehicles (UAVs) in real-world applications such as delivery, inspection, and surveillance. Existing learning-based approaches often degrade under covariate shift and offer limited transparency, making their decisions difficult to interpret and validate on resource-constrained platforms. We present NeuroSymLand, a neuro-symbolic framework for marker-free UAV landing site safety assessment that explicitly separates perception-driven world modeling from logic-based safety reasoning. A lightweight segmentation model incrementally constructs a probabilistic semantic scene graph encoding objects, attributes, and spatial relations. Symbolic safety rules, synthesized offline via large language models with human-in-the-loop refinement, are executed directly over this world model at runtime to perform white-box reasoning, producing ranked landing candidates with human-readable explanations of the underlying safety constraints. Across 72 simulated and hardware-in-the-loop landing scenarios, NeuroSymLand achieves 61 successful assessments, outperforming four competitive baselines, which achieve between 37 and 57 successes. Qualitative analysis highlights its superior interpretability and transparent reasoning, while deployment incurs negligible edge overhead. Our results suggest that combining explicit world modeling with symbolic reasoning can support accurate, interpretable, and edge-deployable safety assessment in mobile systems, as demonstrated through UAV landing site assessment.

cs.RO↗
arXiv2026

The FLAMINGO simulations data release

We describe the public release of $>2.3$ petabytes of data from the FLAMINGO cosmological simulations. The suite consists of hydrodynamical simulations that include radiative cooling, star formation, stellar mass loss and the resulting chemical enrichment, supernova feedback, and two implementations of AGN feedback. Neutrinos are simulated explicitly using particles. Data products include snapshots, halo and galaxy catalogues, HEALPix all-sky lightcone maps, particle data for lightcone maps, and power spectra. The FLAMINGO set includes 22 hydrodynamical simulations. In addition, there are 16 gravity-only simulations, including the $10080^3$ particles FLAMINGO-10k run, with initial conditions that match those of the corresponding hydrodynamical runs. The fiducial hydrodynamical simulations span three numerical resolutions that have each been calibrated to reproduce the present-day galaxy stellar mass function and gas fractions in low-redshift clusters. Other simulations systematically vary the galaxy stellar mass function, cluster gas fractions, cosmology (including neutrino masses), and/or the nature of dark matter, in volumes of 1Gpc$^3$. The release includes hitherto unpublished simulations that use extra dark matter particles. While we provide a facility for downloading complete simulation outputs, we recognise that for many users this will not be possible due to limited local storage or network bandwidth. We implement a web service that enables users to explore available outputs and selectively download datasets or parts of datasets.

astro-ph.CO↗
arXiv2026

"Hadron-in-fat-jet'' AI Tagging to Detect Rare Decays Such as $W^{\pm}\toπ^{\pm}γ$

We investigate a novel class of boosted-object signatures at the LHC, where a high-$p_{\text{T}}$ fat-jet contains an identifiable hadron or quarkonium state originating from rare or semi-exclusive decays. Unlike conventional boosted jet studies, which focus on multi-prong partonic substructure, our approach probes hybrid configurations such as $W^{\pm}\toπ^{\pm}γ$, where a localized hadronic or quarkonium signal is embedded within a collimated jet. By fine-tuning the signature-oriented, pre-trained Sophon AI model optimized for large-radius jets, and combining it with an event-level BDT and a soft-drop-mass shape fit, we obtain an expected 95\% CL upper limit of ${\cal B}(W^{\pm}\toπ^{\pm}γ)<2.78\times10^{-5}$ for $450\,\mathrm{fb}^{-1}$ in our nominal setup. This study serves as a first proof-of-principle demonstration of the ``hadron-in-fat-jet'' paradigm; substantial gains in sensitivity are expected from improved trigger strategies, additional production channels, and dedicated taggers, while the methodology itself is broadly applicable to a wide range of rare Standard Model processes and searches for light or exotic resonances at present and future collider experiments.

hep-ph↗
arXiv2026

FLAMINGO: The thermal history of the Universe from tSZ effect cross-correlations and its dependencies on cosmology and baryon physics

The cross-correlation between tracers of large-scale structure, such as galaxies or quasars, and the thermal Sunyaev-Zel'dovich (tSZ) signal yields a measure of the bias-weighted mean electron pressure, $\langle b_\mathrm{h} P_\mathrm{e} \rangle$, where $b_\mathrm{h}$ is the halo bias and $P_\mathrm{e}$ is the electron pressure. With a model for the bias, one can derive the thermal history, $\mathrm{d}y/\mathrm{d}z$, where $y$ is the Compton parameter and $z$ is redshift. We explore how these quantities depend on redshift, cosmology, and the physics of galaxy formation using the FLAMINGO suite of cosmological hydrodynamical simulations, which spans a range of cosmological parameters and baryonic feedback implementations in volumes of up to $(2.8\,\text{Gpc})^3$. We find that $\langle b_\mathrm{h} P_\mathrm{e} \rangle$ depends steeply on $S_8 \equiv σ_8\sqrt{Ω_\mathrm{m}/0.3}$, with an effective scaling $\langle b_\mathrm{h} P_\mathrm{e} \rangle \propto S_8^{ε(z)}$, where the exponent $ε(z) \approx 3$ over the redshift range $0.1 \leq z \leq 1$. Compared with existing cross-correlation measurements using tracer samples from SDSS, BOSS, eBOSS, DES, and DESI cross-correlated with tSZ measurements from Planck, we find that models with a low-$S_8$ cosmology and strong feedback are preferred, with a joint fit yielding $S_8 = 0.72^{+0.03}_{-0.03}$ and a normalised group-mass halo baryon fraction $f_b(10^{13}\,M_\odot, z=0.1)/(Ω_b/Ω_m) = 0.10^{+0.09}_{-0.05}$ . Contrary to most probes of feedback which sample smaller scales (e.g., X-ray measurements), we show that feedback boosts $\langle b_\mathrm{h} P_\mathrm{e} \rangle$, thus providing a novel test of feedback models. Overall, our results show the thermal history provides a route to jointly constrain cosmological parameters and test models of galaxy formation.

astro-ph.CO↗
arXiv2026

Constraints on SMEFT operators from $Z \to μμbb$ decay

The Standard Model Effective Field Theory (SMEFT) provides a systematic framework to probe indirect effects of heavy new physics via precision measurements. While SMEFT constraints have been extensively studied using purely leptonic $Z$ decays and inclusive $Z$ production, mixed leptonic-hadronic modes remain largely unexplored. In this work, we analyze $Z \to μμbb$ decays within the SMEFT framework, deriving constraints on dimension-six operators that affect four-fermion interactions between leptons and bottom quarks, as well as $Z$-fermion couplings. Signal and background events are simulated with state-of-the-art Monte Carlo tools, including detector effects such as $b$-tagging, and limits on the relevant Wilson coefficients are extracted using kinematic distributions and a profile likelihood approach. Our results provide complementary constraints to existing SMEFT studies and yield the first process-specific limits on flavor-resolved four-fermion operators involving muons and bottom quarks from $Z$ decays.

hep-ph↗
arXiv2026

Beyond Isolation: A Unified Benchmark for General-Purpose Navigation

The pursuit of general-purpose embodied agents is hindered by fragmented evaluation protocols that isolate navigation skills and fixate on specific robot morphologies, failing to reflect real-world scenarios where agents must orchestrate diverse behaviors across varying embodiments. To bridge this gap, we introduce OmniNavBench, a benchmark for cross-skill coordination and cross-embodiment generalization. OmniNavBench introduces three paradigm shifts: (1) Compositional Complexity. We propose composite instructions that interleave sub-tasks from 6 categories (PointNav, VLN, ObjectNav, SocialNav, Human Following and EQA), compelling agents to transition between exploration, interaction, and social compliance within a single episode. (2) Morphological Universality and Sensor Flexibility. We present a simulation platform that breaks the reliance on single-morphology evaluation, enabling generalization tests across humanoid, quadrupedal, and wheeled robots, with a modular sensor interface and 170 environments blending synthetic assets with real-world scans. (3) Demonstrations Quality. Moving beyond shortest-path algorithms, we curate 1779 expert trajectories via human teleoperation, capturing behavioral nuances such as exploratory glance and anticipatory avoidance. Extensive evaluations demonstrate that current methods, despite their claimed unified design, struggle with the complex, interleaved nature of general-purpose navigation. This exposes a critical disparity between existing capabilities and real-world deployment demands, underscoring OmniNavBench as a testbed for the next generation of generalist navigators. Dataset, code, and leaderboard are available at http://omninavbench.cloud-ip.cc.

cs.RO↗
arXiv2026

From Particles to Perils: SVGD-Based Hazardous Scenario Generation for Autonomous Driving Systems Testing

Simulation-based testing of autonomous driving systems (ADS) must uncover realistic and diverse failures in dense, heterogeneous traffic. However, existing search-based seeding methods (e.g., genetic algorithms) struggle in high-dimensional spaces, often collapsing to limited modes and missing many failure scenarios. We present PtoP, a framework that combines adaptive random seed generation with Stein Variational Gradient Descent (SVGD) to produce diverse, failure-inducing initial conditions. SVGD balances attraction toward high-risk regions and repulsion among particles, yielding risk-seeking yet well-distributed seeds across multiple failure modes. PtoP is plug-and-play and enhances existing online testing methods (e.g., reinforcement learning--based testers) by providing principled seeds. Evaluation in CARLA on two industry-grade ADS (Apollo, Autoware) and a native end-to-end system shows that PtoP improves safety violation rate (up to 27.68%), scenario diversity (9.6%), and map coverage (16.78%) over baselines.

cs.SE↗
arXiv2026

Self-consistent secondary cosmic microwave background anisotropies and extragalactic foregrounds in the FLAMINGO simulations

Secondary anisotropies in the cosmic microwave background (CMB) contain information that can be used to test both cosmological models and models of galaxy formation. Starting from lightcone-based HEALPix maps and catalogues, we present a new set of mock CMB maps constructed in a self-consistent manner from the FLAMINGO suite of cosmological hydrodynamical simulations, including CMB lensing, thermal and kinetic Sunyaev-Zeldovich effects, cosmic infrared background, radio point source and anisotropic screening maps. We show that these simulations reproduce a wide range of observational constraints. We also compare our simulations with previous predictions based on dark matter-only simulations which generally model the secondary anisotropies independently from one another, concluding that our hydrodynamical simulation mocks perform at least as well as previous mocks in matching the observations whilst retaining self-consistency in the predictions of the different components. Using the model variations in FLAMINGO, we further explore how the signals depend on cosmology and feedback modelling, and we predict cross-correlations between some of the signals that differ significantly from those in previous mocks. The mock CMB maps should provide a valuable resource for exploring correlations between different secondary anisotropies and other large-scale structure tracers, and can be applied to forecasts for upcoming surveys.

astro-ph.CO↗