Every paper checked — nothing hidden

All papers

Every paper checked for brief #4 of 12–19 Oct 2026 — nothing is hidden. Brief papers first.

All17,313 In brief24 More (on-topic)3,333 Not relevant13,754 Not assessed87 Older352 Could not confirm93 Blogs with no paper266 All blogs284

Paper	Topic	Relevance
Sharper Frames, Worse Rooms: Viewpoint Coverage Dominates Image Quality in Feed-Forward 3D Reconstruction This paper shows that wide viewpoint coverage produces significantly better 3D room reconstructions than selecting only sharp, motion-free camera frames. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Studies geometric frame selection for 3D reconstruction without multimodal models.
Sharper Frames, Worse Rooms: Viewpoint Coverage Dominates Image Quality in Feed-Forward 3D Reconstruction This paper shows that wide viewpoint coverage produces significantly better 3D room reconstructions than selecting only sharp, motion-free camera frames. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Analyzes handheld indoor room 3D reconstruction, not industrial visual inspection.
VIEWCRAFT: Evolving What Multimodal Agents See and How They Read It This paper proposes a framework that jointly evolves visual evidence presentation and reading procedures to improve multimodal agent task performance. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Studies general multimodal agent skills, not industrial defect detection.
Frequency-Step Cache: Separating Synchronization and Repair for Efficient Autoregressive Video Diffusion This paper introduces a training-free caching method that separates cache synchronization and computation repair to accelerate autoregressive video diffusion generation. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Concerns video generation diffusion caching rather than multimodal text-image-video understanding.
Frequency-Step Cache: Separating Synchronization and Repair for Efficient Autoregressive Video Diffusion This paper introduces a training-free caching method that separates cache synchronization and computation repair to accelerate autoregressive video diffusion generation. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on generative video diffusion backbones, not industrial visual inspection.
Frequency-Step Cache: Separating Synchronization and Repair for Efficient Autoregressive Video Diffusion This paper introduces a training-free caching method that separates cache synchronization and computation repair to accelerate autoregressive video diffusion generation. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Introduces a caching algorithm rather than an evaluation benchmark.
Frequency-Step Cache: Separating Synchronization and Repair for Efficient Autoregressive Video Diffusion This paper introduces a training-free caching method that separates cache synchronization and computation repair to accelerate autoregressive video diffusion generation. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on caching performance rather than safety or reliability guardrails.
When Trackers Fail: VLM-Guided Verification and Recovery for Robust Video Object Segmentation This paper presents a verification and recovery framework using vision-language models to correct tracking errors and mask drift in video object segmentation. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses general video object segmentation, not industrial quality or defect inspection.
Representational Difference Distillation Proposes a knowledge distillation method that focuses learning signals on specific structural disagreements between teacher and student representations. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on general representation distillation on CIFAR, not industrial vision.
Representational Difference Distillation Proposes a knowledge distillation method that focuses learning signals on specific structural disagreements between teacher and student representations. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Applies knowledge distillation to vision classification without multimodal interaction.
Representational Difference Distillation Proposes a knowledge distillation method that focuses learning signals on specific structural disagreements between teacher and student representations. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Proposes a distillation loss formulation rather than benchmarking methodologies.
Representational Difference Distillation Proposes a knowledge distillation method that focuses learning signals on specific structural disagreements between teacher and student representations. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Does not address model safety, alignment, or deployment guardrails.
Meaningful Moments: A Corpus of VLM-Derived Temporal Importance Labels for Video Action Recognition Introduces a corpus of vision-language model generated importance scores to improve temporal sampling in video action recognition. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Studies human action recognition datasets, not industrial defect or quality control.
Removing the Compute Bottleneck for Event-based Visual Place Recognition with Spiking Neural Networks Implements an aggressively pruned spiking neural network to perform event-based visual place recognition on commodity CPUs in microseconds. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on single-modality event-based sensing rather than multimodal models.
Removing the Compute Bottleneck for Event-based Visual Place Recognition with Spiking Neural Networks Implements an aggressively pruned spiking neural network to perform event-based visual place recognition on commodity CPUs in microseconds. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses event-based visual place recognition for mobile robotics, not industrial manufacturing.
Hybrid Context Forests and Bilateral Viterbi Matching for 3D Rib Centerline Tracing in CT Presents an open-source framework using context forests and Viterbi dynamic programming to extract and number rib centerlines in CT scans. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses medical CT rib tracing, not industrial visual inspection.
Hybrid Context Forests and Bilateral Viterbi Matching for 3D Rib Centerline Tracing in CT Presents an open-source framework using context forests and Viterbi dynamic programming to extract and number rib centerlines in CT scans. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Exclusively targets 3D medical CT image processing without multimodal reasoning.
From Failure Prediction to Component Selection in Semantic Segmentation Analyzes failure prediction and component selection methods to evaluate which semantic segmentation predictions to retain or inspect. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Studies failure prediction in 2D image semantic segmentation, not multimodal models.
From Failure Prediction to Component Selection in Semantic Segmentation Analyzes failure prediction and component selection methods to evaluate which semantic segmentation predictions to retain or inspect. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on autonomous driving scene segmentation, not industrial defect detection.
PIGEONS: Incremental Long-Video Captioning with Persistent Identity Gallery and Evolving Overarching Narrative Summary Develops an incremental long-video captioning method using a persistent identity gallery and evolving narrative summary. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses general long-video captioning rather than industrial visual inspection.
Regime-Aware Router Learning for Multi-Domain Medical Image Generation with Diffusion Experts Evaluates router-learning strategies for combining domain-specialized diffusion experts in multi-domain medical image generation. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on medical image generation across domains, not industrial inspection.
Regime-Aware Router Learning for Multi-Domain Medical Image Generation with Diffusion Experts Evaluates router-learning strategies for combining domain-specialized diffusion experts in multi-domain medical image generation. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Presents a routing framework for diffusion experts, not model evaluation methods.
Regime-Aware Router Learning for Multi-Domain Medical Image Generation with Diffusion Experts Evaluates router-learning strategies for combining domain-specialized diffusion experts in multi-domain medical image generation. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Does not address deployment safety, hallucination, or adversarial robustness.
Regime-Aware Router Learning for Multi-Domain Medical Image Generation with Diffusion Experts Evaluates router-learning strategies for combining domain-specialized diffusion experts in multi-domain medical image generation. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Involves diffusion image generation rather than cross-modal understanding or VLMs.
Regime-Aware Router Learning for Multi-Domain Medical Image Generation with Diffusion Experts Evaluates router-learning strategies for combining domain-specialized diffusion experts in multi-domain medical image generation. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Studies multi-domain diffusion expert routing rather than small models or efficient inference.
Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding Proposes a training-free frame packing method enabling vision-language models to process long videos with constant memory consumption. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Addresses compute efficiency rather than safety, alignment, or hallucination guardrails.
Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding Proposes a training-free frame packing method enabling vision-language models to process long videos with constant memory consumption. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on general long video VLM inference, not industrial inspection.
Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding Proposes a training-free frame packing method enabling vision-language models to process long videos with constant memory consumption. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Proposes a video compression technique rather than benchmark or evaluation methodology.
Geometry-Guided Semantic Reconstruction for 3D Instance Segmentation Introduces masked semantic reconstruction and geometry modulation to better align lifted 2D semantic priors with 3D instance representations. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Targets general 3D indoor scene segmentation, not industrial manufacturing environments.
A Concept-informed Interpretable Artificial Intelligence Framework for Breast Ultrasound Diagnosis Develops an interpretable AI framework for breast ultrasound diagnosis incorporating clinical standard descriptors and assessment workflows. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses clinical breast ultrasound diagnosis, not industrial defect or anomaly detection.
ProMoDD: Dataset Distillation for 3D Human Motion via Prototype-Conditioned Diffusion Models Proposes a prototype-conditioned diffusion method to distill large 3D human motion datasets into compact representative subsets. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on human motion action recognition, not industrial vision tasks.
ProMoDD: Dataset Distillation for 3D Human Motion via Prototype-Conditioned Diffusion Models Proposes a prototype-conditioned diffusion method to distill large 3D human motion datasets into compact representative subsets. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Studies 3D skeleton action distillation, not multimodal cross-modality models.
All-St3R SLAM: Real-time RGB Dynamic SLAM with Pixel-wise Motion Masking Presents a real-time RGB dynamic SLAM framework that removes moving objects at pixel precision using motion discrepancy analysis. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Addresses geometric dynamic SLAM tracking without multimodal language integration.
All-St3R SLAM: Real-time RGB Dynamic SLAM with Pixel-wise Motion Masking Presents a real-time RGB dynamic SLAM framework that removes moving objects at pixel precision using motion discrepancy analysis. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on autonomous robotics SLAM, not industrial manufacturing inspection.
Esports World Model Dataset Releases a large gameplay dataset across multiple competitive games to support multi-agent interactive world model research. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Introduces an esports video dataset, completely unrelated to industrial vision.
Esports World Model Dataset Releases a large gameplay dataset across multiple competitive games to support multi-agent interactive world model research. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Presents a game-engine video and state dataset rather than multimodal foundation models.
Geometric Parrots: Rule Absence vs. Rule Fragility in Spatial and Physical Reasoning for Vision-Language Models Introduces a procedural benchmark and dual-loop evaluation to test whether vision-language models maintain physical and spatial reasoning rules under pressure. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Evaluates reasoning consistency in VLMs, not industrial visual inspection.
EgoAlign: Situated 3D Understanding without Situation Descriptions Develops an agent-centric 3D language model that grounds spatial understanding in physical poses without requiring text situation descriptions. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Studies 3D question answering for agents, not industrial defect detection.
DIGS: Distribution-Informed Gaussian Splatting for Training-Free Open-Vocabulary 3D Segmentation Introduces a training-free open-vocabulary 3D segmentation approach that maintains per-primitive label distributions for clean object boundaries. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Presents open-vocabulary 3D segmentation, not industrial defect or manufacturing inspection.
VGGA: Geometry-Grounded Autoencoder for Video Generation Proposes a geometry-grounded video autoencoder incorporating structural depth and geometric constraints to improve latent representations for video generation. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on latent geometric autoencoders for video generation, not multimodal models.
VGGA: Geometry-Grounded Autoencoder for Video Generation Proposes a geometry-grounded video autoencoder incorporating structural depth and geometric constraints to improve latent representations for video generation. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Develops a video generation autoencoder, not industrial inspection or defect detection.
Seg2Click: Learning Pixel-Level Action Masks for GUI Visual Grounding Reformulates graphical interface visual grounding as executable-region mask prediction to pinpoint click actions reliably. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Targets GUI agent visual grounding, not manufacturing or industrial anomaly inspection.
MFlowHOI: Whole-Body Human-Object Interaction Generation via Part-Aware MeanFlow Presents a one-step framework to generate whole-body human-object interactions efficiently using structured latent spaces. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on motion generation from text and contacts, lacking broader multimodal reasoning.
MFlowHOI: Whole-Body Human-Object Interaction Generation via Part-Aware MeanFlow Presents a one-step framework to generate whole-body human-object interactions efficiently using structured latent spaces. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Covers human-object interaction motion generation, not industrial visual quality control.
EgoWholeView: Camera-Only Head-Mounted Capture of Body and Bimanual Hand Motion for Embodied Demonstrations Introduces a six-camera head-mounted system to capture body and bimanual hand motions for embodied demonstrations. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on egocentric motion capture for avatars, not industrial visual inspection.
EgoWholeView: Camera-Only Head-Mounted Capture of Body and Bimanual Hand Motion for Embodied Demonstrations Introduces a six-camera head-mounted system to capture body and bimanual hand motions for embodied demonstrations. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on multi-camera motion capture without multimodal language or reasoning components.
Graph3R: Unifying View Graph Initialization with Feed-Forward Reconstruction Combines view graph initialization with feed-forward reconstruction using a shared geometric representation to handle uncurated multi-scene image collections. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Covers multi-view camera geometry and 3D reconstruction, not multimodal understanding.
Graph3R: Unifying View Graph Initialization with Feed-Forward Reconstruction Combines view graph initialization with feed-forward reconstruction using a shared geometric representation to handle uncurated multi-scene image collections. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses 3D multiview scene reconstruction, not industrial defect detection.
FlowAvatar: Real-Time Full-Body Avatars from Sparse Egocentric Inputs on Consumer XR Devices Enables real-time full-body avatar tracking and personalized appearance on standalone consumer headsets using sparse egocentric sensor inputs. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on VR avatar embodiment and pose tracking, not industrial inspection.
FlowAvatar: Real-Time Full-Body Avatars from Sparse Egocentric Inputs on Consumer XR Devices Enables real-time full-body avatar tracking and personalized appearance on standalone consumer headsets using sparse egocentric sensor inputs. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Addresses sensor and camera pose tracking for avatars, not multimodal vision-language models.
DepthTile: Depth-Guided Existing-Image Self-Tiling with Input-Side Tileability Assessment Presents a depth-guided diffusion method to convert images into seamless repeatable tiles while preserving original local geometry and details. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on diffusion-based image tiling using monocular depth, not multimodal reasoning.
DepthTile: Depth-Guided Existing-Image Self-Tiling with Input-Side Tileability Assessment Presents a depth-guided diffusion method to convert images into seamless repeatable tiles while preserving original local geometry and details. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses image self-tiling synthesis, not industrial quality control or inspection.
ProGAN-Based Synthetic Image Generation for Industrial Visual Inspection of Metallic Parts Applies progressive growing generative adversarial networks to generate synthetic images for industrial defect detection on metallic parts. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on GAN-based synthetic image generation, not multimodal vision-language models.
3D Gaussian Splatting for Fine-grained Urban Scene Reconstruction via Semantic-guided Density Control and Loss Design Uses segmentation guidance to improve density control and loss functions for fine-grained 3D Gaussian Splatting scene reconstruction. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Addresses 3D Gaussian Splatting and segmentation, not cross-modal reasoning models.
3D Gaussian Splatting for Fine-grained Urban Scene Reconstruction via Semantic-guided Density Control and Loss Design Uses segmentation guidance to improve density control and loss functions for fine-grained 3D Gaussian Splatting scene reconstruction. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on 3D urban scene reconstruction, not industrial visual inspection.
ClothAtlas: Joint Garment Identification and State Estimation under Self-Occlusion The paper introduces a flow-matching model that jointly identifies garments and estimates full 3D configurations from partial, occluded point clouds. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Studies robotic garment manipulation, not visual inspection or industrial quality control.
ClothAtlas: Joint Garment Identification and State Estimation under Self-Occlusion The paper introduces a flow-matching model that jointly identifies garments and estimates full 3D configurations from partial, occluded point clouds. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on 3D point cloud mesh matching rather than multimodal vision-language models.
Learning Temporal Evidence Selection for Video Anomaly Understanding The paper proposes a learned temporal evidence selection approach to generate grounded text explanations for video anomaly understanding. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on general video anomaly reasoning, not manufacturing or production inspection.
RAR: Rate-Adaptive Rollout for Time-Consistent Interactive World Models The paper presents a rate-adaptive rollout framework that allows interactive video world models to operate consistently across varying temporal rates. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Addresses video generation frame rate adaptation rather than cross-modal understanding.
RAR: Rate-Adaptive Rollout for Time-Consistent Interactive World Models The paper presents a rate-adaptive rollout framework that allows interactive video world models to operate consistently across varying temporal rates. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Covers interactive video world models, not industrial defect detection or visual quality control.

9,841–9,900 of 17,313← Previous Next →