Papers

3D Gaussian Splatting has achieved remarkable success in novel view synthesis. However, extracting high-fidelity surfaces directly from 3DGS remains challenging due to its discrete and unstructured nature. Existing 3DGS-based reconstruction methods typically rely on multi-view geometric consistency or local constraints. Without an explicit structured geometric prior during optimization, these methods often struggle to resolve structural ambiguities, leading to artifacts and floaters, particularly in textureless or occluded regions. To address this limitation, we propose TopoSurfel, a novel framework that closes the loop between Gaussian surfels and continuous meshes. Unlike recent methods that incorporate mesh extraction into the differentiable pipeline by introducing auxiliary neural networks or extra per-Gaussian parameters, we dynamically extract a continuous proxy mesh via a non-trainable differentiable iso-surfacing process. Leveraging this differentiable connection, we introduce a mesh-guided surfel evolution strategy, including normal alignment and geometry-aware density control, to effectively suppress floaters and fill surface holes. Furthermore, to address the initialization challenges in large-scale environments, we propose a spatially aware hybrid re-initialization strategy that ensures robust reconstruction across complex scenes. Extensive experiments demonstrate that TopoSurfel achieves competitive geometric reconstruction accuracy while maintaining high-quality mesh-based novel view synthesis. The code for our method is available at https://github.com/Fan-Treasure/TopoSurfel.

Digital 3D objects used in games and animation are often required to be compositional; that is, decomposed into semantically meaningful parts. Recent 3D generation methods can produce high-quality compositional objects conditioned on image or text prompts. Yet, such global conditioning lacks the precise part-level controllability required for professional creative workflows. To address this, we introduce MultiCube, a novel compositional 3D generation method that provides explicit, independent control over both the semantics and spatial arrangement of each part. MultiCube takes as input a global text prompt, a text schema specifying the desired parts, and a spatial layout indicating the bounding boxes of the parts in the given schema. It outputs a 3D object composed of distinct meshes, one per specified part, that adhere to the given semantic and spatial conditions. Our approach employs a two-stage diffusion process, first generating a schema- and layout-aligned monolithic mesh, then decomposing the mesh into individual parts simultaneously. A novel Part Layout Adapter is used to encode per-part conditions independently of the other parts. Experiments demonstrate that our method can generate high-quality compositional 3D objects with precise part-level control, including those with unique layouts difficult to achieve with text or image prompting alone. Project page: https://multi-cube.github.io

We study how angular energy signals composed of non-negative Monte Carlo path samples can be compressed and reconstructed for irradiance using finite moments. Writing each sample as an energy-weighted directional feature $x = r u$, we adopt total energy, the first directional moment, and the traceless second moment as $1+3+5$ linearly additive, rotationally covariant statistics. Under a fixed Lebesgue reference measure, the maximum-entropy closure yields $p(r,u) \propto \exp(-\beta r g(u))$, where $g(u) = 1 - b \cdot u + u^T Q u$, whose directional probability and angular energy density are proportional to $g^{-3}$ and $g^{-4}$, respectively. When $g_{\min} > 0$ the closure is normalizable and the reconstruction is strictly positive. We further provide analytic moment matching, variance, inverse sampling, and closed-form diffuse response for the pure-dipole four-parameter subfamily, as well as the realizability domain, partition function, azimuthal algebraic integral, and LUT-oriented reconstruction form for the dipole-second-moment coaxial five-parameter subfamily. Experiments cover 981 Poly Haven HDRI 2K scenes and three Debevec probes. Five-parameter MaxEnt achieves a 78.7% per-scene win rate against stored QZH, with mean luminance RMSE reduced by 15.8%; the advantage is more pronounced in scenes with strong directionality. Both MaxEnt variants maintain zero negative irradiance across all scenes. Full second-order SH-2 yields the lowest overall error, while five-parameter MaxEnt ranks second and outperforms SH-2 in the high-directionality bucket; the coaxial subfamily shows systematic closure error on non-coaxial multi-source scenes.

Embodied conversational agents require synchronized full-body motion (body gestures and facial expressions) that aligns with speech and emotional state. Omni-modal large language models excel at multimodal understanding but produce only linguistic outputs, leaving a critical gap in embodied response generation. We identify and address a failure of emotion conditioning: like other conditional generators that under-use weak conditioning signals, a flow-matching model given both a rich audio embedding and a discrete emotion label suppresses the emotion, generating near-identical motion regardless of the specified emotion. We present EMODY Flow, a lightweight (around 35M parameters) flow-matching framework that attaches to a frozen Qwen-3 Omni model and reuses its internal Mimi audio-codecs to condition two parallel DiT generators - one for SMPL-X body pose, one for FLAME facial expressions. A training-time auxiliary emotion classifier restores emotion sensitivity by forcing generated motion to be emotion-identifiable. EMODY Flow sets a new state of the art on BEAT2 gesture quality, with FGD 0.302, Beat Correlation 0.853, and Diversity 24.62 - improving over the best prior results by 26%, 5%, and 62% respectively - and transfers to zero-shot facial animation on TFHP without domain-specific fine-tuning. Beyond these quantitative gains, the classifier yields clearly emotion-separated motion, which we demonstrate qualitatively through a multidimensional-scaling analysis of the generated gestures.

Music-driven character animation enables and enhances transformative applications in entertainment and interactive education. However, synthesizing realistic drumming motion from audio remains challenging due to the inherent tension between high-acceleration dynamics and the need for extreme spatial-temporal precision. Existing approaches, often reliant on motion matching or MIDI input, struggle with generalizing to diverse real-world audio. Moreover, the field lacks standardized evaluation metrics capable of distinguishing precise drumming from noisy motion. In this paper, we introduce a generative diffusion framework featuring a dual-objective loss function that decouples skeletal integrity from drumstick precision, thus enabling centimeter-level stick precision without sacrificing natural body dynamics. Additionally, leveraging our own dataset and data augmentation strategy, the model generalizes to non-curated, in-the-wild audio. To rigorously evaluate performance, we propose two novel metrics: an impact-to-target distance to quantify spatial precision and an audio-motion correlation score to assess temporal alignment. Our quantitative analysis and user studies demonstrate that our system generates high-quality motion that is often indistinguishable from ground-truth performances.

We introduce and analyse an interpolatory subdivision scheme in which each inserted point is defined by a curvature-profile model rather than by an affine combination of its neighbours, and we establish its convergence and geometric regularity. On each edge, curvatures estimated from circumscribed circles prescribe a curvature profile, the segment is reconstructed by Frenet integration with geodesic shooting, and the new point is taken at half arc length. The construction works identically in the plane and on the sphere, reproduces geodesics and circles exactly, and a unique curvature prefilter turns its planar linearisation into the six-point Deslauriers-Dubuc scheme. The straightening condition of Ewald, Reif, and Sabin fails for this scheme. Measuring only the normal part of the relative distortion repairs their theory, and we prove that the refined polygons converge to a regular interpolatory limit curve with continuous tangent direction and curvature, twice continuously differentiable in arc length. Regularity is thus governed by the normal companion scheme alone, as these authors conjectured.

The idea of using Lipschitz extensions [1,2], or Gradually Varied Functions (GVFs)[3], for mesh-free scattered data reconstruction was proposed by the author in 2012 [4]. However, its practical application to modern mesh-free reconstruction methods has not been fully explored. Motivated by recent advances in computational tools, including AI-assisted mathematical programming and software development, we revisit this idea and investigate the use of a Lipschitz extension as an initialization step for Moving Least Squares (MLS) reconstruction [5,6]. Our computational experiments indicate that this initialization significantly improves the stability and reconstruction accuracy of MLS under sparse and irregular sampling. This is a preliminary study intended to establish feasibility; a fuller evaluation with additional benchmarks and comparisons is left to future work.

Programmatic representations provide a compelling paradigm for 3D content creation, enabling fine-grained edits, interpretability, and explicit structural control. Yet, agentic workflows that rely on large language models (LLMs) to author 3D programs remain brittle, often failing to translate high-level intent into consistent low-level geometry. We attribute this fragility to a mismatch between existing programmatic interfaces and the reasoning strengths of LLMs, which favor semantic structure and spatial relations over fragile numeric choices. In this paper, we jointly design an Agent-centric Domain-Specific Language (aDSL) and a role-specialized multi-agent system to close this gap. aDSL bridges semantic logic and geometric constraints by emphasizing composability and spatial reasoning; it enables agents to manipulate geometry through relational operators instead of brittle absolute coordinates. Building on aDSL, our training-free multi-agent system follows a Plan-Execute-Critic loop to decompose requests, synthesize code, and iteratively repair errors and constraint violations using execution feedback. Experiments show that this co-design improves robustness, controllability, and faithfulness to user intent. Our method outperforms prior LLM-based baselines on text-to-shape and image-to-shape tasks while preserving explicit structure, editability, and interpretability. It also enables downstream applications such as articulated object creation and structured scene composition. Our code is available at https://github.com/sig-pku/aDSL.

Point-Based Neural Rendering (PBNR) is emerging as a promising class of rendering techniques, which are permeating all aspects of society, driven by a growing demand for real-time, photorealistic rendering in AR/VR and digital twins. However, achieving real-time PBNR on VR/AR devices is challenging. This paper proposes MetaSapiens v2, a PBNR system that delivers real-time neural rendering on VR/AR devices while maintaining human visual quality. MetaSapiens v2 combines four techniques. First, we present an efficiency-aware pruning technique to optimize rendering speed. Second, we introduce a Foveated Rendering (FR) method with an efficient primitive for PBNR, leveraging humans' low visual acuity in peripheral regions to relax rendering quality and improve rendering speed. Third, we leverage the redundancy between the two eyes and propose a selective warping method to further reduce the computation overhead in AR/VR binocular rendering. Finally, we propose an accelerator design for binocular FR, addressing the load imbalance issue in (FR-based) PBNR and supporting warping for efficient binocular rendering. Our evaluation shows that MetaSapiens v2 achieves an order of magnitude speedup over existing PBNR models while maintaining the visual quality.

Estimating the deformation of solids via physical simulation is an important problem spanning fields such as computer animation, engineering and robotics. Such simulations are computationally expensive and scale poorly when the representation of an object is refined by increasing the level of discretization. Reduced Order Methods (ROM) offer computational savings by decreasing the number of degrees of freedom, for example by using handles that control groups of vertices. We present the first decimation-based algorithm for computing a sparse, compactly supported set of deformation handles. The crux of our method utilizes iterative algebraic simplification to optimize handle deformation to match any input deformation, such as linear vibration modes. This applies to any volumetric input mesh, including those with high genus or porous features, since we do not alter the geometry. We also devise an efficient algorithm to compute and update compact supports and their associated weights. We leverage compact support to develop an efficient, reduced-cubature computation scheme. Once optimized, our handles offer a memory-efficient solution while enabling real-time elastodynamics simulation of complex geometry. We show real-time performance on a variety of tetrahedral meshes with up to 796,623 tetrahedra.

We present the first r-adaptive method for simulating cloth dynamics and statics with frictional contact in modern cloth pipelines. Thin cloth requires high effective spatial resolution to reproduce wrinkles, folds, buckling, and sharp contact features. However, applying existing variational r-adaptivity to piecewise-linear shells reveals two coupled failure modes. Discretized incremental-potential (IP) optimization can become trapped in poor local minima, yielding suboptimal physical configurations. It can also lower IP artificially by collapsing elements, invalidating the finite-element approximation on which the objective relies. We address both problems with degeneracy-activated quality regularization. The regularizer remains inactive for well-shaped elements, preserving anisotropic adaptation and local densification, but becomes strong near degeneracy. It suppresses spurious low-energy basins, improves escape from suboptimal physical minima, and prevents element bunching, a cloth-specific failure in which elements progressively collapse as cloth slides across sharp contact features. For practical performance, we introduce a dynamic nonlinear solver that exploits within-timestep coherence through accelerated derivative evaluation and dynamic IPC tolerance updates for r-adaptive iterative trust-region (ITR) solves. This yields a 3-6x speedup over prior optimal ITR. Experiments on challenging frictional-contact scenarios show that, under equal vertex-count and time-budget constraints, our method achieves higher visual fidelity than fixed meshes.

Terrain level-of-detail is measured almost exclusively on digital elevation models: single-valued, smooth at the sampling scale, sampled from real topography. Game terrain is often none of these. We compare six rendering methods - height-field ray marching, voxel-accelerated ray marching, sliced proxy geometry, per-sample bar rasterization, compute scattering, and a fitted triangle mesh - implemented in a single engine over a single data path, on the hand-authored multi-layer terrain of Vangers (1998), scored against a CPU ray cast of the same source data. Every method must preserve the two solid intervals available at a ground sample, render at interactive rates, and reflect local terrain destruction without reloading the level. These constraints rule out treating caves as decoration or amortizing a static preprocessing step over an immutable map. From the original game's top-down camera the six methods look interchangeable. At eye-level horizons they do not: point scattering loses coverage, slicing bands, and an over-simplified mesh can miss a wall. At the selected quality settings a greedy triangulated irregular network (TIN) has the lowest mean frame time on every device we measured, but the fit cost is set by the second layer rather than by floor relief, and making that mesh editable retains 319 MiB of GPU geometry and 535 MiB of CPU triangulation. All six implementations use the same native wgpu / WebGPU API and canonical WGSL. We release the engine, the harness, and a one-command measurement protocol.

3D Gaussian Splatting has made Gaussian primitives a highly efficient representation for real-time novel view synthesis, but its rasterisation-based formulation relies on screen-space approximations that limit accurate view-dependent ordering and the integration of secondary ray effects such as reflections, refractions, and shadows. Gaussian ray tracing addresses these limitations by evaluating explicit ray-primitive intersections, yet it remains costly to train. We observe that the main bottleneck is not ray traversal alone, but the pixel-centric backward propagation, where many threads concurrently accumulate gradients into the same primitive parameters, causing severe atomic contention and thread serialisation. We present 3DGART, a practical training framework for ray-traced Gaussian rendering. Our key idea is to reorganise backward propagation around primitives rather than pixels. Using conservative perspective-correct screen-space bounds, we build a compact intermediate buffer and a tile-primitive mapping that allows each thread to accumulate the contribution of one primitive over its covered pixels within a tile. This transforms gradient computation from a contention-heavy scatter operation into a structured gather-like process. On Mip-NeRF 360, 3DGART achieves an $\approx 3-3.5\times$ raw training speedup over per-pixel baseline and $\approx4 \times$ over 3DGRT on Mip-NeRF 360 while improving quality. More importantly, 3DGART makes fully ray-traced Gaussian training practical, reaching runtimes competitive with rasterisation-based pipelines while preserving benefits of ray tracing.

We propose a remeshing-enhanced method for generatively deforming shapes with visual losses. It is intuitive that sufficiently drastic deformations of a mesh without changing its triangulation can easily compromise element quality, even if such large geometry changes may be semantically desired. Shape deformation methods could thus benefit from changing the triangulation; however, this is not done by most generative, text-based, visually-supervised mesh deformation methods. Remeshing is a discrete operation, proven to be especially challenging to couple with the notoriously noisy supervision signal provided by visual losses. We propose a vertex-based deformation optimization quantity capable of large deformations and robustness to such noise; we periodically remesh using an isotropic remesher that interpolates and carries forward the deformation optimization state. This enables continuous, geometry-informed progress in coarse-to-fine addition of resolution. The resulting shapes' triangulations fit their optimized geometry and have neat isotropic elements. Further, our method is localizable, able to grow new features on a base shape with expressive detail, leaving the rest unchanged. We showcase the effectiveness of our method on a variety of shapes and prompts, both local and global deformations, and demonstrate its superior visual quality and triangle efficiency. Our project page is at https://threedle.github.io/radmesh.

The massive overactuation in the human musculoskeletal system makes it challenging to train musculoskeletal models to generate human-like motion via reinforcement learning, primarily because exploration in the resulting high-dimensional and redundant action space is extremely inefficient. To address this problem, we propose the $\lambda$-hold controller, inspired by the equilibrium-point (EP) hypothesis, which has been widely supported by extensive evidence from human motor control studies. The policy's control variable is the per-muscle EP threshold length $\lambda$, from which a stretch-reflex recruitment law computes the muscle excitations automatically. Holding each $\lambda$ over an interval of the gait phase also sharply reduces the frequency at which the policy must be queried. Consequently, the controller, to our knowledge for the first time, enables a muscle-actuated skeletal model to learn human-like sprinting using only a minimal reward within an hour of training. The efficient exploration through the proposed $\lambda$-hold controller is not merely an engineering trick but an approach grounded in physiology, bringing together the EP hypothesis, intermittent control, and optimal feedback control. Beyond encapsulating human-like behavior in predictive simulation, this achievement contributes to developing a learnable model of the human motor controller.

Recent monocular depth estimators achieve strong zero-shot generalization, yet often struggle to preserve fine-grained structures and object boundaries. We attribute this limitation to the prevalent combination of large-patch ViT encoders and convolutional decoders, as coarse tokenization can weaken pixel-level cues that upsampling cannot fully recover. To address this issue, we propose PXDepth, a discriminative monocular depth model that separates global context modeling from pixel-level depth prediction. Specifically, a large-patch ViT captures global scene context, while a pixel-space predictor composed of Context-Modulated Pixel Transformer blocks maintains high-resolution spatial representations throughout depth estimation. This design preserves fine structures and sharp boundaries without sacrificing global depth consistency. Across diverse zero-shot benchmarks, PXDepth combines faithful local geometry with competitive global depth accuracy while remaining efficient at inference. Our code and model are available at https://yuanzhy29.github.io/PXDepth-Page/.

Differentiable triangle-soup reconstruction inherits a limitation from its objective: photometric and geometric losses cannot measure topology, so a reconstruction with a collapsed loop or a punctured enclosed void can score exactly as well as a correct one (on Chamfer-equal probes the diagrams differ 35-40x in bottleneck distance). The standard implicit remedy – steer *where* the resampler spends its budget – does not repair this: in a controlled study, a topology-informed prior is largely matched by an equally wide random one, and no prior shape repairs loops. We therefore move topology into the objective: a differentiable persistence term compares the evolving surface's diagram, measured on live surface samples, to a fixed target; gradients flow through a pair-frozen backward re-expressing matched birth/death simplices as closed-form circumradii, plus a recruitment term restoring the gradient optimal matching provably lacks when a feature is missing; one ratio knob calibrates the loss against the photometric gradient, no curriculum needed. Every claim passes a channel-controlled verdict: the loss must beat a norm-matched *non-topological* control through the identical gradient channel, at Chamfer parity. Under that rule the loss is topology-specific for enclosed voids (4.0-7.9x lower error) and – the class every allocation prior failed – for loops (2.3x, zero phantom handles, while the control collapses one); loss and prior compose; component counts (H0) are a null result. The verdicts replicate without per-shape tuning on eight external genus-known meshes in two pre-registered groups (group means: loops 1.52x, voids 4.87x; the one non-pass is a no-headroom null), degrading gracefully under noise. All evidence is synthetic and single-machine, with the target diagram known in advance; real scans are future work. Prescription: correct topology in the loss, allocate wide, combine.

Per-pixel rendering difficulty is conventionally characterised by one noisy scalar: the sample variance of a Monte Carlo estimator. We argue that it should instead be characterised through transport structure — a discrete description of how each contribution's energy reaches the sensor, deterministic under stated renderer conventions — with variance treated as a measurement whose reliability that structure helps predict. We make this concrete with a seven-class transport-mechanism descriptor assigned per contribution event, evaluate it on eleven scenes, and measure variance reliability on the seven first-party ones. The dominant label agrees $87$–$99.6\%$ between 64 and 4096 samples per pixel, where quantile-binned variance agrees as little as $21\%$; its stability on unseen scenes is predicted from their pilots. Conditioning a pilot variance on the label improves equal-budget sample allocation wherever heavy-tailed buckets carry appreciable population, reduces to the incumbent where they are absent or negligible, and is neither reproduced by a random partition nor absorbed by a median-of-means estimator; where the pilot fails for other reasons, as on the classical ajar-door scene ($6.8$~dB below uniform), the label says so from the pilot alone.

One of the main reasons compute hardware becomes obsolete is software bloat: resource requirements increase for every iteration of a software product. Resource constrained VMs are one way to combat software bloat as they post a hard limit on the resources and so force the programmer to be frugal. In this paper we explore the deployment of one such resource constrained VM, Uxn, on GPU. We show that for competitive performance it is essential to make use of the GPU data parallelism. We present an OpenMP-style parallelism API for Uxntal, the stack-based assembly-style language for the Uxn platform. We demonstrate that exemplar code using our API can run at comparable performance even on an integrated GPU. Specifically, our evaluation results show that using this approach improves performance on the compute-intensive Stencil benchmark with 19x and frame rate on the graphics-intensive Bunnymark benchmark with 7x. In practice, all laptops and desktops and even mobile devices have a GPU and our work shows that they can be used to execute frugal workloads effectively.

Identity-preserving video generation (IPVG) requires synthesizing videos that are faithful to both reference subjects and text prompts. Existing methods are often hindered by high tuning costs or limited input-level enhancements, struggling to maintain rigid identity consistency during complex, long-sequence actions. To address these limitations, we propose KeyID, a training-free IPVG framework that decouples the synthesis of video dynamics from the injection of identity. Specifically, KeyID comprises two components: (1) Reference-Aware Video Generation, which produces an identity-agnostic video draft aligned with multiple references, and (2) Identity-Preserved Keyframe Editing, which integrates the target identity via sparse keyframe correction and subsequent motion interpolation. By shifting from dense frame-level supervision to sparse keyframe-level refinement, KeyID effectively resolves the capacity conflict between prompt adherence and identity fidelity. Crucially, our modular design allows seamless extension to multi-subject references and complex sequential action generation without additional training. KeyID outperforms prior works and is validated by automatic and human evaluations on the official challenge benchmark, ultimately securing the runner-up position in the Track 2 (Sequential Action) of the ACM Multimedia 2026 IPVG Grand Challenge. Source code is available at https://github.com/WISLab-GDUT/KeyID.

We present AnyTalk, a novel method for generating 3D speech animations for arbitrary characters without requiring any animation data. While existing audio-driven 3D speech animation methods rely on character-specific training data or laborious rigging/re-meshing, AnyTalk circumvents these limitations by leveraging recent video diffusion models trained on extensive video datasets. We first adapt a pre-trained video diffusion model to a target character through our Character-specific Fine-tuning (CsF) technique. By fine-tuning on rendered images of the 3D character paired with zeroed-out audio embeddings (representing "no motion"), we eliminate the need for animation data while preserving the motion prior of large-scale video diffusion model. We then uplift the resulting talking-head video into a 3D speech animation by estimating blendshape parameters through a proposed optimization process. AnyTalk enables lip-synced animations across diverse face meshes and blendshape configurations, significantly reducing manual effort and data requirements. We further enhance usability by distilling AnyTalk into a streamlined network, $\text{AnyTalk}_{RT}$, thereby enabling real-time performance. By leveraging talking-head video generation, our method broadens access to audio-driven speech animation technology for arbitrary characters. The code is publicly available at https://serin-yoon.github.io/projects/anytalk/.

Interactive video generation and editing are becoming increasingly important for creative design. In this report, we introduce EditStream: a unified framework for interactive video generation and editing. EditStream unifies multiple video creation and manipulation tasks within a single DiT-based model through flexible task-specific conditioning, and further transforms it into a fast, few-step autoregressive model for efficient streaming. It supports Text-to-Video, Image-to-Video, Video-to-Video, Editing Propagation, Reference-guided Video Editing, and Camera Pose Change, enabling flexible control over video generation, transformation, and editing within one system. To make the unified model practical for interactive use, we develop a two-stage distillation approach that combines Velocity Moment Matching (VMM) with autoregressive unrolling. VMM matches conditional velocity moments at student-reached intermediate states to preserve generation quality and motion, while unrolling exposes the student to its own autoregressive predictions to improve temporal stability. Together, they alleviate common challenges in few-step autoregressive video generation, including over-saturation, degraded motion, temporal instability, and complex training. EditStream provides a practical and scalable solution that bridges high-quality diffusion-based video models with interactive creative workflows.

Different shape representations facilitate different computations. Surface representations, in particular meshes, are often used for modeling, whereas volume representations are useful for spatial queries such as intersection or containment. Optimizing a surface representation based on a volumetric properties by gradient descent requires the derivatives of the volume relative to its bounding surface. We derive this gradient for winding numbers and show that it can be efficiently computed for volumetric values sampled on a regular grid (voxel representation) and surface parameters based on vertex sets (triangle meshes). This enables an efficient solution for a variety of optimization problems. We demonstrate the practical use of this approach at the examples of deforming meshes to resolve intersections, being manufacturable by cutting with a bandsaw from three directions, and creating shapes that are close to tiling 3D space.

Implicit surface representations have regained popularity because of their use in machine learning. A common component in optimization is regularization, penalizing the deviation of the surface from its original shape. The popular as-rigid-aspossible (ARAP) energy strikes a good compromise between realistic deformation behavior and efficient computation, at least for piecewise linear meshes. We develop an approach for computing the ARAP energy of a deformation function based on point sampling of the surface. The implicit representation is exploited to provide differentials in each sample. The evaluation is efficient and exact in each sample (up to numerical precision). We demonstrate the general applicability of the method to neural shape processing in several applications and contrast its properties with alternatives from the literature.

Learning to generate or reconstruct explorable worlds requires video paired with more than RGB: camera motion, scene geometry, temporal correspondence and, for interactive models, control signals. Real capture can provide some of these signals, but dense geometry and long-range correspondence usually rely on estimation or specialised instrumentation. Rendering provides these quantities directly, yet existing synthetic resources rarely combine them on the same frames while also supporting controlled changes of viewpoint and appearance. We introduce WorldRover, a data engine for generating richly annotated, long-range explorations of artist-built environments. At its core, WorldRover-Engine is an Unreal Engine pipeline that executes and offline-renders minute-scale routes while preserving their full trajectories and scene geometry. The same exploration can be replayed from first-person, third-person, and 360 panoramic cameras under different environmental states. Using WorldRover-Engine, we construct WorldRover-10M, whose sequences pair RGB with metric depth, camera trajectories, and trajectory-derived action signals throughout each exploration. Third-person subsets additionally provide dense optical flow, long-range 2D/3D point tracks with visibility, and a character trajectory distinct from the camera trajectory. The engine can render a traversal from first-person, third-person and 360 panoramic viewpoints, under different environmental states or with a neutral white material, while preserving the route and scene geometry. WorldRover therefore turns long-horizon world exploration into a scalable data-generation problem, providing supervision for models that must build, maintain, and revisit coherent representations of an explorable world.

Functional connectome analysis examines brain-region interactions to understand and identify disorders such as autism spectrum disorder and Alzheimer's disease. Existing methods typically use GNNs and Transformers to model the full functional connectivity matrix. However, processing tens of thousands of connections introduces redundancy and noise, increases computational cost, and limits connection-level interpretability. This raises a central question: do we really need complex interaction modeling, or is identifying a small set of disease-relevant connectivity patterns sufficient? To answer this question, we propose BrainLinear, a lightweight geometry-aware framework for mining disease-discriminative connectome patterns. BrainLinear first maps each functional connectivity matrix to a shared tangent space centered at the Fréchet mean of the training set, capturing subject-specific deviations while respecting matrix geometry. It then scores each ROI-pair tangent direction by its classification contribution and disease–control difference, retaining Top-$K$ directions as a compact representation. Finally, a shallow multilayer perceptron performs classification on the selected representation. Experiments on ABIDE and ADNI show that BrainLinear matches or exceeds strong GNN and Transformer baselines at a fraction of their cost: it improves AUC and ACC over the best baseline for each metric by up to $3.54$ and $1.39$ percentage points, while reducing runtime and peak GPU memory by $84.0\%$ and $68.4\%$ relative to the closest baseline in AUC. The selected directions are directionally consistent with between-group displacements and organized across major functional systems, supporting connection-level interpretation.

Joint audio–video (AV) generators are trained on data in which what an event looks like and what it sounds like are spuriously correlated. We present a controlled causal study of the resulting failure mode. In an AV structural causal model where the audio is, by construction, independent of the video's nuisance appearance, models that let audio read video directly—through cross-attention or a shared latent—learn a visual shortcut: they predict sound from appearance rather than the causal event and, when the appearance–event correlation is broken at test time, synthesize the wrong event's sound. Crucially, the popular remedy of routing both modalities through a shared common-cause latent does not fix this—a bottleneck, an unsupervised shared/private factorization, and a faithful shared-prior model all grab the appearance proxy and fail like the direct model. Blocking the shortcut instead requires an intervention on the nuisance: under the stated assumptions we prove that counterfactual invariance is necessary and sufficient to identify the causal predictor, and we verify the mechanism from feature-vector SCMs to procedural pixel video, real images with spectrogram audio, moving real digits, and a conditional generator. On a real, pretrained V2A generator (MMAudio), an input-intervention test shows the model is far from invariant to sound-irrelevant edits, though a generic-noise control reveals it is broadly input-brittle rather than specifically colour-shortcutting—clean isolation of the shortcut needs the controlled confounds our synthetic studies provide. We characterize when the shortcut occurs, compare the objective against supervised counterfactual augmentation, and isolate the unknown-nuisance regime—where the intervention cannot be applied—as the central open problem.

Part-aware 3D object generation is essential for graphics applications such as controllable modeling, editing, and articulation, where objects are represented as coherent assemblies of semantic parts. However, existing part-aware generation methods, do not scale well to highly complex objects. As the number of parts increases, generating detailed geometry becomes prohibitively expensive in token length and memory. We introduce MegaParts, a scalable autoregressive 3D generation framework to address this challenge by combining structured sequence modeling with a token-efficient vector-quantized shape tokenizer. Our tokenizer learns discrete latent representations for part-level geometry by minimizing token usage subject to high-fidelity reconstruction, enabling adaptive-length tokenization based on geometric complexity. On top of this compact representation, we train a large language model to generate object bounding boxes, part bounding boxes, and part shape tokens within a unified structured sequence. Combined with efficient long-context training strategy, our token-efficient formulation scales to objects with up to 300 parts and sequence lengths up to 256k tokens. This substantially extends the scale of part-aware 3D generation while preserving compositional structure and enabling fine-grained part-level control. Our method achieves higher mesh quality than baseline autoregressive and diffusion models, showing that compressed discrete part tokens improve not only scalability but also the achievable fidelity of generated geometry. These results suggest that LLM native token-efficient autoregressive modeling is a compelling alternative to diffusion for large-scale part-aware 3D generation. The project page is available at https://expmaster.github.io/megaparts_webpage.

We present an uncertainty-aware Jacobi set computation method. In general, Jacobi sets are topological descriptors that capture the gradient alignments of two scalar fields, as, e.g., used for multi-field visualization. We adopt and reformulate an existing computational approach that relies on an edge-based identification of Jacobi set edges on a given triangulation. Our extension to uncertainty visualization builds upon a versatile, spatially coherent uncertainty model for pairs of scalar fields based on multivariate normal distributions. We propagate the uncertainty analytically, thereby lifting the original Jacobi set computation to uncertain inputs. Furthermore, we present an overlay of visual mappings specifically designed to show the Jacobi sets along with different facets of uncertainty information. Both the uncertainty model and uncertainty-aware method are validated against a Monte Carlo approach on an analytic dataset and applied to two use cases from fluid dynamics and weather ensembles.

Lagrangian modeling is vital to fluid dynamics, as it characterizes particle transport and complements the Eulerian representation. However, Lagrangian trajectories are less commonly available than Eulerian fields, while most neural operators are trained and evaluated primarily in the Eulerian representation. This mismatch motivates a new learning problem: can a model trained solely on Eulerian observations generalize zero-shot from Eulerian field prediction to Lagrangian particle rollout, without Lagrangian supervision or task-specific adaptation? To address this problem, we propose the Transferable Latent Operator (TLO), which learns a unified flow representation shared by Eulerian field prediction and Lagrangian particle rollout. TLO decouples latent flow evolution from coordinate-dependent decoding: querying the evolving latent representation at fixed spatial coordinates yields Eulerian fields, whereas querying velocities at particle positions and recursively updating these positions enables Lagrangian rollout. Across five fluid-dynamics benchmarks, TLO consistently outperforms existing neural operators in both Eulerian field prediction and zero-shot Lagrangian rollout, with further gains from limited Lagrangian fine-tuning.