{"total":22,"items":[{"citing_arxiv_id":"2605.21489","ref_index":50,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Variance Reduction for Expectations with Diffusion Teachers","primary_cat":"cs.LG","submitted_at":"2026-05-20T17:59:52+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"CARV amortizes upstream diffusion teacher costs over noise resamples with timestep importance sampling and stratified-inverse-CDF sampling, delivering 2-3x effective compute gains in text-to-3D experiments and order-of-magnitude variance cuts in single-step distillation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.18052","ref_index":165,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Efficient 3D Content Reconstruction and Generation","primary_cat":"cs.CV","submitted_at":"2026-05-18T08:41:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Presents Instant3D for rapid text/image-to-3D generation via multi-view diffusion plus feed-forward reconstruction, and FastMap for 10x faster structure-from-motion with comparable accuracy.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"lems for outlier removal and global position recovery. ShapeFit and ShapeKick[73] provide robust and scalable formulations for global translation estimation. 11 Bundle adjustment is the gold standard of camera pose refinement and is a major com- putational bottleneck. Scaling strategies include large-scale problem formulations[4], out-of- core optimization[165], multicore parallelization[271], and software packages that implement generic sparse bundle adjustment[144]. 2.2.2 Learnable Components in Traditional SfM Pipelines Classical SfM pipelines decompose reconstruction into multiple stages, including feature extraction, feature matching, geometric verification, and multi-view optimization, etc. Many of these steps now have learned counterparts that can be used as drop-in replacements."},{"citing_arxiv_id":"2605.11913","ref_index":18,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Vector Scaffolding: Inter-Scale Orchestration for Differentiable Image Vectorization","primary_cat":"cs.CV","submitted_at":"2026-05-12T10:27:30+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Vector Scaffolding uses Interior Gradient Aggregation, Progressive Stratification, and Rapid Inflation Scheduling to achieve 2.5x faster optimization and up to 1.4 dB higher PSNR in differentiable vectorization.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"primitives such as 2D Gaussians [6,10] or tetrahedral meshes [7]. Beyond 3D scene representation, Gaussian-based primitives have also been explored for 2D image modeling. Works such as GaussianImage [31] and Image- GS [32] demonstrate that collections of 2D Gaussians can serve as compact and expressive alternatives to raster images or implicit neural representations [18,21]. Vector Scaffolding 5 Fig.2: Overview of Vector Scaffolding.(a)Interior Gradient Aggregation: Opti- mization is stabilized by aggregating internal area gradients alongside boundary gradi- ents via the Reynolds transport theorem. (b)Rapid Inflation Scheduling:Progressive Stratificationaligns vector representation with the natural power law of image fre-"},{"citing_arxiv_id":"2605.09312","ref_index":9,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Low-Cost Neural Radiance Fields","primary_cat":"cs.CV","submitted_at":"2026-05-10T04:13:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"Comparative study of DS-NeRF, TensoRF, and HashNeRF with depth-supervision and architectural variants finds no conclusive outperformance under equal training time but identifies which design choices transfer to low-data, low-compute regimes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09071","ref_index":8,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Probability-Flow Distillation: Exact Wasserstein Gradient Flow for High-Fidelity 3D Generation","primary_cat":"cs.CV","submitted_at":"2026-05-09T17:27:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Probability-Flow Distillation exactly matches the Wasserstein gradient flow of the target distribution when distilling 2D diffusion priors into 3D models, yielding higher-fidelity results than SDS or SDI.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.08824","ref_index":12,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"HairGPT: Strand-as-Language Autoregressive Modeling for Realistic 3D Hairstyle Synthesis","primary_cat":"cs.GR","submitted_at":"2026-05-09T09:19:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"HairGPT reframes 3D hairstyle synthesis as dual-decoupled autoregressive strand sequence modeling with geometric tokenization for semantic control and rare style generation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.21400","ref_index":27,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"You Only Gaussian Once: Controllable 3D Gaussian Splatting for Ultra-Densely Sampled Scenes","primary_cat":"cs.CV","submitted_at":"2026-04-23T08:07:42+00:00","verdict":"UNVERDICTED","verdict_confidence":"MODERATE","novelty_score":6.0,"formal_verification":"none","one_line_summary":"YOGO reformulates stochastic 3D Gaussian Splatting into a deterministic budget-aware system and supplies an ultra-dense dataset to enforce physical fidelity over viewpoint interpolation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.18083","ref_index":29,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Implicit neural representations as a coordinate-based framework for continuous environmental field reconstruction from sparse ecological observations","primary_cat":"cs.LG","submitted_at":"2026-04-20T10:59:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Implicit neural representations enable stable, resolution-independent reconstruction of continuous environmental fields from sparse and irregular ecological data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.17414","ref_index":45,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Physics-Aware Query-Conditioned Graph Attention Networks for Radio Map Estimation","primary_cat":"eess.SP","submitted_at":"2026-04-19T12:44:29+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A physics-aware query-conditioned hierarchical graph attention network estimates point-wise transmitter-resolved radio maps from sparse measurements and outperforms baselines on DeepMIMO simulations in direct, residual, and gated regimes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.13476","ref_index":56,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"RobotPan: A 360$^\\circ$ Surround-View Robotic Vision System for Embodied Perception","primary_cat":"cs.RO","submitted_at":"2026-04-15T04:58:23+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"RobotPan predicts metric-scaled compact 3D Gaussians from calibrated multi-view inputs via spherical coordinates and hierarchical voxel priors for real-time 360° robotic perception and reconstruction.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.12626","ref_index":17,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Habitat-GS: A High-Fidelity Navigation Simulator with Dynamic Gaussian Splatting","primary_cat":"cs.RO","submitted_at":"2026-04-14T11:52:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Habitat-GS integrates 3D Gaussian Splatting scene rendering and Gaussian avatars into Habitat-Sim, yielding agents with stronger cross-domain generalization and effective human-aware navigation.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"and natively integrating drivable gaussian avatars, uniquely combining photo- realistic rendering, dynamic high-fidelity humanoids, and a mature open-source research ecosystem. 2.2 Neural Rendering Neural Radiance Fields (NeRF) [16] demonstrated that implicit neural scene representations can synthesize photorealistic novel views from multi-view im- ages.Subsequentworkssignificantlyimprovedtrainingspeedandrenderingqual- ity [2,17]. However, NeRF's volume rendering paradigm requires per-pixel ray marching, yielding frame rates far below the real-time requirements of embodied Habitat-GS 5 Table 1: Comparison of Embodied AI simulation platforms.\"GPU Req.\" in- dicates hardware requirements beyond standard CUDA-capable GPUs. \"No Special Req.\" means standard GPUs are sufficient."},{"citing_arxiv_id":"2604.11172","ref_index":17,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"NeuVolEx: Implicit Neural Features for Volume Exploration","primary_cat":"cs.GR","submitted_at":"2026-04-13T08:30:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"NeuVolEx extracts robust spatial features from INR training via a structural encoder and multi-task scheme to enable accurate ROI classification with limited supervision and unsupervised viewpoint clustering in volume exploration.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"sentation by feeding 3D coordinates into a multiresolution hash grid Fig 3. Method Overview Hash Grid Encodingγ𝛽 FiLMMulti-HeadLearning Scheme IntensityGradient Local StdLocal Mean ++ Structural EncodingMain MLP 3D Coordinates(X, Y, Z) Structural Patches(𝑛!voxels) Fig. 3: An architectural overview of our volume exploration-optimized INR. encoding [17]. This encoding maps each voxel to trainable features at multiple spatial resolutions. Another pathway samples a local patch of n3 voxels (where n=5 ) centered at each coordinate from the raw volume. This patch is processed by a structural encoder with two hid- den layers of 32 channels to generate a structural representation. By encoding local neighborhood around each voxel, this pathway provides"},{"citing_arxiv_id":"2604.09543","ref_index":60,"ref_count":3,"confidence":0.9,"is_internal_anchor":false,"paper_title":"ANTIC: Adaptive Neural Temporal In-situ Compressor","primary_cat":"cs.LG","submitted_at":"2026-04-10T17:58:16+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ANTIC reduces storage for large-scale PDE simulations by orders of magnitude through adaptive temporal snapshot selection combined with continual neural-field residual compression while preserving physics accuracy.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.08945","ref_index":22,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TouchAnything: Diffusion-Guided 3D Reconstruction from Sparse Robot Touches","primary_cat":"cs.CV","submitted_at":"2026-04-10T04:26:36+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TouchAnything reconstructs accurate 3D object geometries from only a few tactile contacts by optimizing for consistency with a pretrained visual diffusion prior.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16266","ref_index":23,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Patchwork: A compact representation for 3D polygonal shapes","primary_cat":"cs.GR","submitted_at":"2026-03-24T15:20:32+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Patchwork is a new compact shape representation for 2D and 3D geometry that approximates arbitrary shapes with arbitrary precision using a small number of parameters, provable bounds, and gradient-based optimization with pruning regularization.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.14199","ref_index":22,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Learnable Multi-level Discrete Wavelet Transforms for 3D Gaussian Splatting Frequency Modulation","primary_cat":"eess.IV","submitted_at":"2026-02-15T15:49:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Multi-level DWT frequency modulation in 3DGS reduces Gaussian counts by recursive low-frequency decomposition and a single scaling parameter while preserving rendering quality.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2602.04516","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TACO: Temporal Consensus Optimization for Continual Neural Mapping","primary_cat":"cs.RO","submitted_at":"2026-02-04T13:07:08+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TACO reformulates neural implicit mapping as temporal consensus optimization to enable continual adaptation to scene changes without data replay or storage.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2508.01014","ref_index":20,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Hestia: Voxel-Face-Aware Hierarchical Next-Best-View Acquisition for Efficient 3D Reconstruction","primary_cat":"cs.RO","submitted_at":"2025-08-01T18:27:23+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Hestia improves generalizable next-best-view planning for 3D reconstruction via hierarchical action search, diverse data, close-greedy strategy, and face-aware voxel design, yielding higher coverage and lower Chamfer distance than prior RL-based methods.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2403.15651","ref_index":22,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"GaNI: Global and Near Field Illumination Aware Neural Inverse Rendering","primary_cat":"cs.CV","submitted_at":"2024-03-22T23:47:19+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"GaNI combines NeuS geometry reconstruction with a light-position-aware inverse neural radiosity stage that adds implicit near-field modeling, surface angle loss, and roughness smoothness priors to recover reflectance parameters from co-located light-camera captures.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2311.04400","ref_index":22,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"LRM: Large Reconstruction Model for Single Image to 3D","primary_cat":"cs.CV","submitted_at":"2023-11-08T00:03:52+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"LRM is a large transformer that predicts a NeRF directly from a single image after training on a million-object multi-view dataset.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2307.03017","ref_index":38,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"RealLiFe: Real-Time Light Field Reconstruction via Hierarchical Sparse Gradient Descent","primary_cat":"cs.CV","submitted_at":"2023-07-06T14:31:01+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RealLiFe optimizes multi-plane images with HSGD to deliver real-time light field reconstruction from sparse views, claiming 100x speedup over offline methods and 2 dB PSNR gain over online ones.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2210.00379","ref_index":42,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"NeRF: Neural Radiance Field in 3D Vision: A Comprehensive Review (Updated Post-Gaussian Splatting)","primary_cat":"cs.CV","submitted_at":"2022-10-01T21:35:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"A literature survey of NeRF and neural field methods from 2020-2025, organized by architecture and application taxonomies with benchmarks and dataset overviews, covering both pre- and post-Gaussian Splatting periods.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}