Search arXiv⌕ Search

arXiv subjects

Yang Zhao

Publications and source records attributed to Yang Zhao.

At least 19 recordsLinked to original sources
arXiv2026

Rényi Multi-entropy for Stabilizer States: A Finite Field Gauss Sum Problem

Rényi multi-entropy probes multi-partite structure beyond bipartite spectra, but its behavior in general quantum states and the entanglement structure it captures remain poorly understood. We develop a finite field Gauss sum framework for calculating it in stabilizer states. Within this framework, the replica contraction reduces to a quadratic Gauss sum if and only if the state is locally unitarily equivalent to a stabilizer state. We obtain analytic expressions for the Rényi multi-entropy of all tree graph states and large families of loop graph states. In contrast to their flat bipartite Rényi spectra, stabilizer states generically have nonflat Rényi multi-entropy spectra.

hep-th↗
arXiv2026

IKMo: Image-Keyframed Motion Generation with Trajectory-Pose Conditioned Motion Diffusion Model

While text-to-motion generation is mainstream, integrating explicit spatial constraints has become a standard solution for precise animation. Existing state-of-the-art models are tailored for professional workflows, effectively processing high-quality, pre-aligned 3D inputs via unified conditioning mechanisms. However, this strictly aligned setting restricts accessibility for general users who rely on unaligned, in-the-wild 2D reference images. This accessible scenario introduces a fundamental conflict: heterogeneous geometric constraints, where user-defined global trajectories clash with local image-derived poses. To resolve this coordinate mismatch, we propose TPMDM, a diffusion model featuring a decoupled parallel control strategy that explicitly isolates and aligns these divergent signals. Furthermore, we introduce IKMo, a holistic framework that synergizes TPMDM with a VLM-powered multi-agent system. By bridging the semantic-geometric gap, IKMo translates user-friendly inputs into explicit motion constraints to drive the diffusion backbone. Experiments on HumanML3D and KIT-ML demonstrate that our method establishes a new state-of-the-art in the task of generating motions under heterogeneous trajectory and keyframe pose constraints.

cs.GR↗
arXiv2026

OmniTrace: A Unified Framework for Generation-Time Source Attribution in Omni-Modal LLMs

An omni-modal language model can answer a question about a multimodal document, a recording, or a video, but it cannot show which part of the input the answer came from. Asking the model to cite its evidence yields what it says it used; measuring what it used has so far required rerunning it many times with parts of the input removed, a cost that grows with every second of audio and video. We show that the generation pass already holds most of this information. OmniTrace attributes each generated sentence to source units (text chunks, images, and seconds of audio or video) by reading late-layer attention as mass over units with no extra forward pass and no per-modality tuning; head weights can be trained where ablation data are available. With fixed causal test setting, the units OmniTrace selects are causally important for responses and for individual sentences in video, audio, and image-text question answering on Qwen3-Omni-30B, MiniCPM-o 4.5 and Qwen2.5-Omni-7B. With one generation pass, OmniTrace recovers 77-96% of the effect of leave-one-out attribution, which needs up to 41 extra passes. OmniTrace also agrees best with human-marked evidence: its response-level rankings reach an AUROC of 0.83 on documents and 0.64 on video, against 0.76 and 0.60 for the strongest baseline.

cs.CL↗
arXiv2026

SceneReVis: A Self-Reflective Vision-Grounded Framework for 3D Indoor Scene Synthesis via Multi-Turn RL

Current one-pass 3D scene synthesis methods often suffer from spatial hallucinations, such as collisions, due to a lack of deliberative reasoning. To bridge this gap, we introduce SceneReVis, a vision-grounded self-reflection framework that employs an iterative "diagnose-and-act" loop to explicitly intercept and resolve spatial conflicts using multi-modal feedback. To support this step-wise paradigm, we construct SceneChain-12k, a large-scale dataset of causal construction trajectories derived through a novel reverse engineering pipeline. We further propose a two-stage training recipe that transitions from Supervised Fine-Tuning to Agentic Reinforcement Learning, evolving the model into an active spatial planner. Extensive experiments demonstrate that SceneReVis achieves state-of-the-art performance in high-fidelity generation and goal-oriented optimization, with improved transfer to underrepresented room categories in the dataset and promising potential for extension to out-of-distribution room types.

cs.CV↗
arXiv2026

Holographic multipartite entanglement dynamics in AdS$_3$-Vaidya

We study how multipartite entanglement is dynamically reorganized during holographic thermalization following a global quench in AdS$_3$/CFT$_2$. We first use the $n$-partite information $(-1)^n I_n$ to probe collective multipartite entanglement in holographic configurations where the full $n$-region entanglement wedge is connected while all fewer-party ones are disconnected, thereby excluding fewer-party contributions. The spatial range of multipartite entanglement first expands and then contracts as the system approaches its late-time locally thermal state. Entanglement involving different numbers of parties develops on comparable early-time scales, while the entanglement that involves more parties relaxes more slowly, revealing a transient propagation from shorter to longer spatial distances. We further compute the Markov gap and the genuine tripartite multi-entropy as complementary probes of tripartite entanglement. The Markov gap can remain enhanced after local thermalization, whereas the genuine tripartite multi-entropy undergoes a nonmonotonic evolution and returns to its vacuum value for the adjacent tripartition considered in this work. These results show that a global quench redistributes the entanglement across spatial scales and reorganizes its multipartite structure.

hep-th↗
arXiv2026

PWM: Personalized World Models with Online Reinforcement Learning

Pretrained world models can generate diverse environments, yet users often want to explore a particular scene specified by their own video. This requires learning the scene's visual identity while retaining the quality of action-conditioned generation. We introduce Personalized World Models (PWM), a framework for customizing interactive world models from short scene videos through online reinforcement learning. In PWM, the support trajectory and its associated controls provide reward feedback on continuations sampled from the current policy. In the GRPO instantiation, group-relative optimization updates a compact LoRA adapter using a unified reward for scene appearance, visual continuity, and motion, while base-policy anchoring regularizes changes to the pretrained generation prior of a frozen Yume-5B backbone. The same adaptation procedure is applied across real and rendered environments. We also instantiate PWM with DiffusionNFT as an alternative reward-guided optimization method for learning the scene-specific adapter. We also introduce PWM-Bench, comprising 150 customization tasks across Indoor, Outdoor, and Gaming, with paired evaluation on held-out continuations. The GRPO and DiffusionNFT instantiations of PWM improve customization over native Yume in 71.3% and 65.3% of the evaluated scenes, respectively, with positive mean gains across all three domains. For the GRPO instantiation, matched SFT comparisons further demonstrate higher mean customization gains and better mean image-quality scores in every domain, while retaining frame-level visual quality close to the pretrained model.

cs.CV↗
arXiv2026

A Dependency-Aware Multi-GPU Training Framework for Temporal Graph Neural Networks

Temporal graph neural networks (TGNNs) achieve superior performance in many dynamic graph learning tasks. These models use node state vectors (i.e., memories) to capture and record graph changes (i.e., events), and then exploit these memories to assist prediction. However, this memory-centric design makes multi-GPU training challenging because predictions for later events often depend on memory updates from earlier events. To maintain model quality, existing multi-GPU TGNN training schemes strictly preserve all event dependencies, which limits parallelism and causes expensive cross-GPU memory synchronization. In this work, we observe that many of these dependencies can be handled more flexibly without harming training quality. Specifically, (a) dependencies in TGNN training have high locality in both time and graph topology, (b) the short-range dependencies missed within a large batch can often be recovered cheaply, and (c) many memory updates are not immediately needed. Based on these observations, we propose DepBloom to accelerate multi-GPU TGNN training with three techniques: (1) temporally bounded spatial partitioning to reduce unnecessary cross-GPU synchronization and improve parallelism; (2) intra-batch memory refinement to recover missing short-range temporal dependencies inside enlarged batches; and (3) critical-path optimization to prioritize urgent memory updates while deferring non-critical work during multi-GPU pipeline execution. Experimental results across representative models and diverse datasets show that DepBloom achieves an average speedup of 2.84$\times$ and up to 5.16$\times$ over state-of-the-art methods, while maintaining comparable or better model accuracy.

cs.LG↗
arXiv2026

On-Chip Shaping of Surface Acoustic Waves via Continuous Diffractive Modulation

Precise shaping of surface acoustic waves (SAWs) on anisotropic piezoelectric substrates is complicated by elastic anisotropy and electromechanical coupling. Here we introduce a continuous diffractive acoustic lens (CDAL), in which weak phase-velocity perturbations produced by subwavelength metallization, together with diffraction, drive lateral field redistribution. An anisotropic thin-element propagation model enables efficient inverse design of the CDAL contour. Laser Doppler vibrometry verifies prescribed lateral field profiles and multichannel beam splitting with unequal widths. By combining continuous diffractive modulation with discrete source-phase encoding, we further shape the lateral field into a Bessel profile, demonstrating joint amplitude-and-phase control. This work provides a high-precision and fabrication-compatible mechanism for on-chip SAW field modulation.

physics.class-ph↗
arXiv2026

Erasing and Regenerating the Turbulent Cascade Arrow

Which information tells a turbulent flow which way to cascade? We apply Haar-distributed Hermitian phases that retain prescribed same-mode quadratic information, including modal amplitudes, the energy spectrum, and same-mode two-time correlations. For the phase-Haar ensemble, the expectation of any integrable odd flux functional is zero. Finite ensembles strongly suppress mean transfer and forward-event bias in the primary block. An independent block retains sign separation but fails the scale-8 mean criterion. In matched evolution protocols, nonlinear Navier-Stokes dynamics restores the selected signed-flux scalar close to its natural reference, whereas linear evolution and an amplitude-matched frozen control do not. Flux and a local cubic geometry scalar recover together, yet high local correlation can survive erasure of the arrow. Signed transfer thus requires information beyond the retained quadratic record; the tested nonlinear dynamics rebuilds it.

physics.flu-dyn↗
arXiv2026

Code2Games: Enabling Coding Agents for Gaming World Generation

Generating a high-quality gaming world from a natural-language game intent requires joint reasoning about scene structure, spatial layout, gameplay objectives, interactive entities, and executable gameplay logic. Existing coding agents can generate individual assets, scenes, or scripts, but often struggle to maintain consistency across these components. We propose Code2Games, an agentic framework that builds a structured gaming world upon a base Blender world generated from the same game intent. Code2Games coordinates scene analysis, gameplay planning, constrained gaming-world generation, and gaming-engine customization through a shared scene-gameplay representation with persistent element correspondence. After world generation, Code2Games adapts the generated world to Unreal Engine 5 and employs an execution-guided reconstruction process that uses compilation diagnostics, runtime feedback, and gameplay test results to resolve inconsistencies arising during engine adaptation. To systematically evaluate gaming-world generation, we introduce the GameCode4D benchmark, which comprises ten fixed game prompts spanning different levels of scene and gameplay complexity. We evaluate the generated results across four dimensions: visual quality, interactive fidelity, multimodal artifact quality, and playable-game quality. Experiments demonstrate that, compared with direct gaming-world generation by coding agents and existing baseline methods, Code2Games consistently improves the visual quality and interactive fidelity of generated gaming worlds, as well as the quality of the resulting games after engine adaptation.

cs.CV↗
arXiv2026

LSR-Ben: A Logical and Scientific Reasoning Benchmark for Evaluating Process Reward Models

Currently, process reward models (PRMs) have exhibited remarkable potential for test-time scaling. Since large language models (LLMs) regularly generate flawed intermediate reasoning steps when tackling a broad spectrum of reasoning and decision-making tasks, PRMs are required to possess capabilities for detecting process-level errors in real-world scenarios. However, existing benchmarks primarily focus on mathematical reasoning, thereby failing to comprehensively evaluate the error detection ability of PRMs across diverse reasoning scenarios. To mitigate this gap, we introduce LSR-Ben, a process-level benchmark specifically designed for assessing PRM's performance across two primary reasoning domains (scientific and logical reasoning) and nine subdomains. We conduct extensive experiments on a diverse set of 22 models, encompassing both PRMs and LLMs, and derive two key findings: (1) In domains beyond mathematical reasoning, the error-detection ability of existing PRMs and LLMs is found to be markedly weaker by comparison. (2) In general, LLMs exhibit a tendency toward over-identification of errors compared to PRMs, whereas PRMs exhibit an inherent tendency to overlook errors compared to LLMs. We hope LSR-Ben can foster future researches on PRMs for broader domains, thereby enhancing the reasoning capabilities of LLMs.

cs.AI↗
arXiv2026

Mara Chain: Rethinking Failure as a Stepping Stone for AI System Auto-Evolution

Optimizing deployed AI systems increasingly amounts to editing prompts, skills, harnesses, and code rather than model weights. Existing approaches commonly optimize these artifacts through propose-evaluate-select procedures, where candidate configurations are evaluated and only those meeting an acceptance criterion are selected. Yet our analysis shows that discarded candidates often contain information critical for subsequent optimization. Discarding them causes later proposals to revisit the same failure modes. We introduce Mara Chain, a refinement procedure that turns rejected candidates into stepping stones. Rather than discarding a rejected candidate, Mara Chain retains and iteratively refines it using evidence accumulated across preceding attempts. The procedure limits each refinement chain to a fixed depth and applies Pareto-filtered Top-N selection to bound the candidate pool. Across AppWorld skill optimization, TerminalBench 2.1 harness optimization, and MuSiQue retrieval-pipeline optimization, Mara Chain delivers greater task-performance gains with fewer rollouts. It outperforms GEPA, ACE, and SkillOpt-Lite by up to 20.5% in relative performance on AppWorld, reaching the target score with 65.5% fewer rollouts than GEPA. It improves the pass rate by 20.2 and 22.5 percentage points over AHE and Meta-Harness on TerminalBench 2.1, respectively, and improves MuSiQue test nDCG@10 and Recall@10 by 0.104 and 0.131 over a hand-written retrieval pipeline.

cs.LG↗
arXiv2026

Learning Informative Prior with Infinite-Dimensional Continuous Normalizing Flow for Bayesian Inverse Problem

This paper addresses infinite-dimensional Bayesian inference for inverse problem of partial differential equations with model parameters in infinite-dimensional Hilbert space. To effectively incorporate prior information, we propose a novel continuous normalizing flows based infinite-dimensional model. Specifically, by introducing a well-defined neural ordinary differential equation in infinite-dimensional space, a simple reference measure can be transformed into a more complex measure which encodes the prior information. A corresponding theoretical framework is established to ensure the well-posedness of our proposed Bayesian prior in infinite-dimensional space. We also provide training methods of the prior for two distinct data settings, along with two sampling algorithms for the resulting Bayesian posterior. The proposed framework is applied to three representative inverse problems: the simple smooth inverse problem, inverse scattering problem, and the inverse heat conduction problem. Numerical experiments support the theoretical analysis and demonstrate the efficiency of the proposed algorithms.

math.NA↗
arXiv2026

CAffNet: Hard Constraint-Affine Neural Networks

We present a novel framework for embedding hard constraint satisfaction into neural network (NN) architectures, specifically feedforward neural networks and transformers, with input-dependent affine constraints of arbitrary cardinality. Traditional constraint enforcement approaches either rely on penalty-based soft constraints, which offer no guarantee of satisfaction, or on post-processing methods that enforce constraints after the NN is trained, which may lead to suboptimality. We introduce a trainable constraint-affine (CAffine) layer into NNs, yielding CAffNet, which goes beyond enforcing affine constraints via fixed orthogonal or parallel projections and enables joint optimization with network parameters. Moreover, we impose no restrictions on the constraint space dimensions and establish that our construction preserves the universal approximation properties of NNs, while providing provable guarantees on constraint adherence for all inputs. Experimental validation demonstrates robust performance across diverse domains requiring guaranteed constraint satisfaction. The code is available at https://github.com/ice-t-lab/CAffNet.git.

cs.LG↗
arXiv2026

Learning Safe-by-Design Neural Network Controllers

Safety filters constructed from control barrier functions (CBFs) are commonly appended to pre-trained neural network controllers to enforce safety requirements. However, this decoupled design with hand-tuned, fixed CBF parameters often fails to adapt to the underlying controller, yielding overly conservative solutions. Thus, given a valid CBF, we address these limitations by jointly learning a neural network controller and neural-network-parameterized CBF parameters, enforcing the resulting affine safety constraints by construction and avoiding an online quadratic program (QP) safety filter at run time. To further improve computational efficiency and scalability, we introduce a lightweight projection architecture that enforces constraints without full constraint enumeration. Extensive simulation evaluations demonstrate reliable, scalable safety constraint satisfaction at reduced computational cost.

eess.SY↗
arXiv2026

EgoPathBench: Evaluating Zero-Shot Egocentric Waypoint Decision-Making in Vision-Language Models

Zero-shot waypoint navigation requires vision-language models to select, from the current first-person observation, a sequence of spatial actions that is feasible for the agent and reaches the goal, placing joint demands on the integrated spatial intelligence of today's foundation VLMs. Existing spatial-intelligence benchmarks primarily evaluate isolated judgments of relations, directions, or targets and therefore do not directly measure the integrated navigation ability required to combine target recognition, action-consequence assessment, distance estimation, and path planning. To fill this evaluation gap, we introduce EgoPathBench, a dataset and five-task benchmark for first-person waypoint decision-making. Each question presents an egocentric RGB image, a natural-language goal, and numbered visible waypoints; a model returns traversable candidates or an ordered route. Predictions are evaluated for candidate feasibility, adjacent-edge legality, and goal arrival under point-agent or embodied geometry. EgoPathBench contains 31,852 training, 1,345 validation, and 1,111 benchmark questions and retains at least one geometrically verified reference route for every route question. Across nine VLMs, the highest EgoPath Score is only 28.3. The top-ranked model reaches 35.9% success on Point Path, but only 2.9% and 4.0% on Embodied Path and Intent Path, respectively, showing that current models remain limited in forming complete, goal-consistent routes under embodiment constraints. Beyond the evaluation data, we release the corresponding training resource. Fine-tuning Qwen 3.5 4B on the released training split raises its EgoPath Score from 3.9 to 38.9 and improves all four reported evaluations across three external spatial benchmarks, with gains of 1.4--9.6 points.

cs.CV↗
arXiv2026

SegKAN: High-Resolution Medical Image Segmentation with Long-Distance Dependencies

Hepatic vessels in computed tomography scans often suffer from image fragmentation and noise interference, making it difficult to maintain vessel integrity and posing significant challenges for vessel segmentation. To address this issue, we propose an innovative model: SegKAN. First, we improve the conventional embedding module by adopting a novel convolutional network structure for image embedding, which smooths out image noise and prevents issues such as gradient explosion in subsequent stages. Next, we transform the spatial relationships between Patch blocks into temporal relationships to solve the problem of capturing positional relationships between Patch blocks in traditional Vision Transformer models. We conducted experiments on a Hepatic vessel dataset, and compared to the existing state-of-the-art model, the Dice score improved by 1.78%. These results demonstrate that the proposed new structure effectively enhances the segmentation performance of high-resolution extended objects. Code will be available at https://github.com/goblin327/SegKAN

eess.IV↗
arXiv2026

PathoHR: Breast Cancer Survival Prediction on High-Resolution Pathological Images

Breast cancer survival prediction in computational pathology presents a remarkable challenge due to tumor heterogeneity. For instance, different regions of the same tumor in the pathology image can show distinct morphological and molecular characteristics. This makes it difficult to extract representative features from whole slide images (WSIs) that truly reflect the tumor's aggressive potential and likely survival outcomes. In this paper, we present PathoHR, a novel pipeline for accurate breast cancer survival prediction that enhances any size of pathological images to enable more effective feature learning. Our approach entails (1) the incorporation of a plug-and-play high-resolution Vision Transformer (ViT) to enhance patch-wise WSI representation, enabling more detailed and comprehensive feature extraction, (2) the systematic evaluation of multiple advanced similarity metrics for comparing WSI-extracted features, optimizing the representation learning process to better capture tumor characteristics, (3) the demonstration that smaller image patches enhanced follow the proposed pipeline can achieve equivalent or superior prediction accuracy compared to raw larger patches, while significantly reducing computational overhead. Experimental findings valid that PathoHR provides the potential way of integrating enhanced image resolution with optimized feature learning to advance computational pathology, offering a promising direction for more accurate and efficient breast cancer survival prediction. Code will be available at https://github.com/AIGeeksGroup/PathoHR.

eess.IV↗