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
Efficient Fréchet Distributional Decoder Alignment for Latent Generative Models Proposes a post-training decoder alignment method for latent generative models to resolve distribution mismatches between training and generation latents. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on image generative alignment rather than small or quantized language models.
CÆNL: Collapse-Based Active Neural Learning with Objective-Aligned Representation Control Investigates active learning dynamics by regularizing representation collapse and managing sample acquisition across training iterations. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant General active learning theory, not industrial visual inspection.
CÆNL: Collapse-Based Active Neural Learning with Objective-Aligned Representation Control Investigates active learning dynamics by regularizing representation collapse and managing sample acquisition across training iterations. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on single-modality vision classification, not multimodal models.
CÆNL: Collapse-Based Active Neural Learning with Objective-Aligned Representation Control Investigates active learning dynamics by regularizing representation collapse and managing sample acquisition across training iterations. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Studies active learning optimization rather than AI deployment safety or robustness.
CÆNL: Collapse-Based Active Neural Learning with Objective-Aligned Representation Control Investigates active learning dynamics by regularizing representation collapse and managing sample acquisition across training iterations. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Focuses on active learning algorithms rather than LLM evaluation benchmarks.
CÆNL: Collapse-Based Active Neural Learning with Objective-Aligned Representation Control Investigates active learning dynamics by regularizing representation collapse and managing sample acquisition across training iterations. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Deals with sample acquisition in active learning, not model compression.
Gaussian Splatting for Real-Time Scene Reconstruction: Methods, Systems, and Trade-offs Surveys real-time 3D Gaussian Splatting scene reconstruction methods, systems, and practical trade-offs across five major technical categories. Peer-reviewed	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Surveys general 3D Gaussian Splatting scene reconstruction, not industrial inspection.
Gaussian Splatting for Real-Time Scene Reconstruction: Methods, Systems, and Trade-offs Surveys real-time 3D Gaussian Splatting scene reconstruction methods, systems, and practical trade-offs across five major technical categories. Peer-reviewed	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Covers 3D novel view synthesis methods rather than multimodal vision-language models.
TwinPrune: Density-Aware Two-Phase Token Pruning for Vision-Language Models Proposes a two-phase token pruning strategy for vision-language models and identifies evaluation flaws in common decode-phase benchmarks. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on vision-language model token pruning, not industrial inspection.
CrossPortrait: Cross-Species Portrait Animation via Adversarial Motion Alignment Introduces a cross-species portrait animation framework that decomposes motion into head poses and expression keypoints to prevent driving-species leakage. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Deals with facial portrait animation rather than industrial visual inspection.
CrossPortrait: Cross-Species Portrait Animation via Adversarial Motion Alignment Introduces a cross-species portrait animation framework that decomposes motion into head poses and expression keypoints to prevent driving-species leakage. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on video face reenactment rather than multimodal foundation models.
Physical-Field-Bridged Material Parameter Inference from Multi-View Deformation Videos Presents a single-pass framework inferring simulation-ready material parameters from multi-view deformation videos using projected physical response fields. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on vision-based physical parameter estimation, not multimodal language models.
Physical-Field-Bridged Material Parameter Inference from Multi-View Deformation Videos Presents a single-pass framework inferring simulation-ready material parameters from multi-view deformation videos using projected physical response fields. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses robotic deformable object manipulation rather than manufacturing quality control.
PLEDGE: Progressive Lossy Error-Bounded Compression of 3D Gaussian Splatting Presents an error-bounded progressive compression codec for 3D Gaussian Splatting with user-declared tolerances for primitive distortion. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on 3D computer graphics compression, not industrial defect inspection.
PLEDGE: Progressive Lossy Error-Bounded Compression of 3D Gaussian Splatting Presents an error-bounded progressive compression codec for 3D Gaussian Splatting with user-declared tolerances for primitive distortion. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Addresses 3D Gaussian Splatting compression, not multimodal models.
Background2Background: a self-supervised background removal framework for high dimensional microscopy Introduces a self-supervised background removal framework for high-dimensional microscopy using local pixel shuffling and noise-reduction techniques. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Addresses specialized optical microscopy imaging rather than multimodal language models.
Background2Background: a self-supervised background removal framework for high dimensional microscopy Introduces a self-supervised background removal framework for high-dimensional microscopy using local pixel shuffling and noise-reduction techniques. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on biological/scientific microscopy rather than manufacturing defect inspection.
Smartphone-Based, Field-Deployed AI Inspection System for Power Distribution Infrastructure Presents a field-deployed smartphone inspection system combining deep learning and geometry to assess power distribution infrastructure conditions. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Applies standard vision techniques to inspection rather than advancing multimodal foundation models.
From Pixels to Executable Physics: Behavior-Grounded World Models for Embodied Action Consequence Prediction The paper presents a physically consistent world model incorporating robot-terrain contact dynamics to predict quadruped action outcomes from an RGB image. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Addresses quadruped robot locomotion, not industrial defect detection.
From Pixels to Executable Physics: Behavior-Grounded World Models for Embodied Action Consequence Prediction The paper presents a physically consistent world model incorporating robot-terrain contact dynamics to predict quadruped action outcomes from an RGB image. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on video world models for robotics rather than multimodal foundation models.
Calibrated Disambiguation for Partial Multi-label Learning Proposes a calibrated disambiguation model to optimize sample weighting when training classifiers on datasets with noisy or excessive labels. Peer-reviewed	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Deals with multi-label learning, not language model evaluation or benchmarks.
Calibrated Disambiguation for Partial Multi-label Learning Proposes a calibrated disambiguation model to optimize sample weighting when training classifiers on datasets with noisy or excessive labels. Peer-reviewed	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on partial-label classification, not deployment safety or guardrails.
Calibrated Disambiguation for Partial Multi-label Learning Proposes a calibrated disambiguation model to optimize sample weighting when training classifiers on datasets with noisy or excessive labels. Peer-reviewed	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Standard classification under label noise without multimodal model focus.
Calibrated Disambiguation for Partial Multi-label Learning Proposes a calibrated disambiguation model to optimize sample weighting when training classifiers on datasets with noisy or excessive labels. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Addresses noisy multi-label training, not model compression or efficient serving.
Calibrated Disambiguation for Partial Multi-label Learning Proposes a calibrated disambiguation model to optimize sample weighting when training classifiers on datasets with noisy or excessive labels. Peer-reviewed	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant General machine learning disambiguation method, not industrial computer vision.
FrameLite: Decision-Oriented Aesthetic Cropping with a Compact Vision-Language Model Proposes a lightweight aesthetic cropping model built on a small vision-language model using multi-stage candidate optimization. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on aesthetic image cropping, not manufacturing or industrial inspection.
MemAct: Streaming Spatial Understanding with Memory and Active Exploration Presents an embodied agent framework combining streaming spatial memory with active exploration for visual reasoning in physical environments. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on embodied spatial navigation and exploration, not industrial visual defect detection.
Code2Worlds-2: Physically Plausible Agent Harness for Embodied World Simulation Extends dynamic world generation to produce physics-executable simulations with collision meshes and physical dynamics for embodied agents. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on compiling 3D/4D visual scenes into physics simulation formats rather than multimodal model capabilities.
Code2Worlds-2: Physically Plausible Agent Harness for Embodied World Simulation Extends dynamic world generation to produce physics-executable simulations with collision meshes and physical dynamics for embodied agents. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on 4D simulation generation for embodied robotics, not industrial manufacturing inspection.
Open-o3-Video-v2: Seek, Ground, and Reason with Verifiable Spatio-Temporal Evidence Develops an agentic video reasoning framework that iteratively identifies missing information, gathers visual evidence, and grounds temporal clues. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on spatio-temporal video question answering, not industrial defect or visual quality inspection.
ScenePercept: Scenario-Conditioned Perceptual Assessment of Real-World User-Generated Content Images Introduces an evaluation benchmark to assess user-generated image quality and aesthetics across diverse real-world scenes. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on image quality and aesthetic assessment, not multimodal vision-language reasoning.
ScenePercept: Scenario-Conditioned Perceptual Assessment of Real-World User-Generated Content Images Introduces an evaluation benchmark to assess user-generated image quality and aesthetics across diverse real-world scenes. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Assesses perceptual quality of user-generated mobile photos, not industrial manufacturing quality control.
Integrating 3D Posture Neutralization with Statistical Shape Modeling for Head-Neck-Shoulder Ergonomic Design Presents a framework integrating posture neutralization with statistical shape modeling to isolate anatomical variations for ergonomic product design. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Deals with 3D scan shape modeling and posture neutralization, not multimodal language models.
Integrating 3D Posture Neutralization with Statistical Shape Modeling for Head-Neck-Shoulder Ergonomic Design Presents a framework integrating posture neutralization with statistical shape modeling to isolate anatomical variations for ergonomic product design. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on 3D anatomical modeling for ergonomic product design, not industrial defect detection.
Deficit-Driven Stratified Acquisition for Building STRATA-S12: An Open SAR-Optical Corpus for Self-Supervised Learning Introduces a deficit-driven stratified acquisition algorithm to create a balanced, co-registered satellite imagery corpus for self-supervised Earth observation. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Presents an acquisition strategy for satellite imagery, not an evaluation framework for LLMs.
Deficit-Driven Stratified Acquisition for Building STRATA-S12: An Open SAR-Optical Corpus for Self-Supervised Learning Introduces a deficit-driven stratified acquisition algorithm to create a balanced, co-registered satellite imagery corpus for self-supervised Earth observation. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on remote sensing dataset collection, not model compression or efficient inference.
Deficit-Driven Stratified Acquisition for Building STRATA-S12: An Open SAR-Optical Corpus for Self-Supervised Learning Introduces a deficit-driven stratified acquisition algorithm to create a balanced, co-registered satellite imagery corpus for self-supervised Earth observation. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Focuses on satellite data collection rather than deployment safety, guardrails, or hallucination.
Deficit-Driven Stratified Acquisition for Building STRATA-S12: An Open SAR-Optical Corpus for Self-Supervised Learning Introduces a deficit-driven stratified acquisition algorithm to create a balanced, co-registered satellite imagery corpus for self-supervised Earth observation. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on Earth observation satellite SAR-optical dataset acquisition, not manufacturing inspection.
Hierarchical Curriculum Learning for Surgical Video Generation: Focusing on Bilateral Axillo-Breast Approach Robotic Thyroidectomy Datasets Adapts text-to-video diffusion models for surgical procedure simulation using hierarchical procedural decomposition and progressive curriculum learning. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on robotic surgical video generation, not manufacturing or industrial visual inspection.
SWORD: Scene-Grounded Wasserstein Object-Relation Distance for 3D Scene Graph Prediction Introduces a scene-grounded Wasserstein distance objective to optimize graph-level semantic proximity in 3D scene graph prediction tasks. Peer-reviewed	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on 3D scene graph prediction losses, not manufacturing or industrial inspection.
SWORD: Scene-Grounded Wasserstein Object-Relation Distance for 3D Scene Graph Prediction Introduces a scene-grounded Wasserstein distance objective to optimize graph-level semantic proximity in 3D scene graph prediction tasks. Peer-reviewed	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Deals with 3D computer vision scene graph prediction rather than multimodal vision-language models.
A Differentiable 3D Scene Graph Metric via Contextual Hellinger Triplet Geometry Proposes a differentiable graph-level loss function to measure semantic and structural discrepancies between predicted 3D scene graphs and ground truth. Peer-reviewed	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on 3D scene graph supervision for generative graphics, not industrial inspection.
A Differentiable 3D Scene Graph Metric via Contextual Hellinger Triplet Geometry Proposes a differentiable graph-level loss function to measure semantic and structural discrepancies between predicted 3D scene graphs and ground truth. Peer-reviewed	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on scene graph loss formulation, not model compression or efficient inference.
A Differentiable 3D Scene Graph Metric via Contextual Hellinger Triplet Geometry Proposes a differentiable graph-level loss function to measure semantic and structural discrepancies between predicted 3D scene graphs and ground truth. Peer-reviewed	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Studies 3D scene graph prediction in computer vision, not multimodal language models.
A Differentiable 3D Scene Graph Metric via Contextual Hellinger Triplet Geometry Proposes a differentiable graph-level loss function to measure semantic and structural discrepancies between predicted 3D scene graphs and ground truth. Peer-reviewed	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Proposes a geometric training loss rather than AI deployment safety, hallucination, or alignment.
A Differentiable 3D Scene Graph Metric via Contextual Hellinger Triplet Geometry Proposes a differentiable graph-level loss function to measure semantic and structural discrepancies between predicted 3D scene graphs and ground truth. Peer-reviewed	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Proposes a training loss for 3D scene graphs, not LLM benchmark evaluation methods.
DT-LoRA: Reaching Beyond the Tangent Space of Low-Rank Adaptation Introduces a low-rank adaptation method that splits rank budgets into stable and dynamic branches to capture residual gradients outside the tangent space. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Studies parameter-efficient fine-tuning optimization rather than AI safety, guardrails, or hallucination prevention.
DT-LoRA: Reaching Beyond the Tangent Space of Low-Rank Adaptation Introduces a low-rank adaptation method that splits rank budgets into stable and dynamic branches to capture residual gradients outside the tangent space. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Focuses on a general parameter-efficient fine-tuning technique rather than multimodal vision-language models.
DT-LoRA: Reaching Beyond the Tangent Space of Low-Rank Adaptation Introduces a low-rank adaptation method that splits rank budgets into stable and dynamic branches to capture residual gradients outside the tangent space. Preprint	LLM evaluation and benchmarks via arxiv · 15 Oct, 06:18	Not relevant Presents a parameter-efficient fine-tuning algorithm rather than an LLM evaluation benchmark or framework.
DT-LoRA: Reaching Beyond the Tangent Space of Low-Rank Adaptation Introduces a low-rank adaptation method that splits rank budgets into stable and dynamic branches to capture residual gradients outside the tangent space. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Proposes a general fine-tuning parameter-efficient technique, not industrial defect or visual quality inspection.
AttenIR: Aligning Dense Visual Attention with Mesh Warping for Unsupervised Image Retargeting Proposes an unsupervised image retargeting framework combining dense visual attention with mesh warping to preserve content without manual annotations. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on image resizing and retargeting, not industrial inspection or quality control.
AttenIR: Aligning Dense Visual Attention with Mesh Warping for Unsupervised Image Retargeting Proposes an unsupervised image retargeting framework combining dense visual attention with mesh warping to preserve content without manual annotations. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Deals with traditional vision mesh deformation for retargeting, not multimodal language models.
DexStitch: Learning and Composing Dexterous Skills from Human Videos via Contact-Aware Robotic Inpainting Presents a framework that segments long-horizon human manipulation videos into skills and translates them into executable robotic actions. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on robotic dexterous manipulation from human video demonstrations, not industrial visual inspection.
DexStitch: Learning and Composing Dexterous Skills from Human Videos via Contact-Aware Robotic Inpainting Presents a framework that segments long-horizon human manipulation videos into skills and translates them into executable robotic actions. Preprint	Multimodal models via arxiv · 15 Oct, 06:18	Not relevant Studies robotic imitation learning and video inpainting, not multimodal vision-language reasoning.
When, Where, and Why: Hierarchical Causal Explainability for Video Face Understanding with MLLMs Introduces an explainability framework and weakly supervised video-text corpus to evaluate causal and temporal facial expression understanding in multimodal models. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on human facial expression understanding in videos, not industrial inspection or defect detection.
MHVF: Training-Free GUI Grounding via Verification-Guided Multi-Hypothesis Search This paper introduces a training-free framework combining multi-hypothesis search and local visual verification to ground interface instructions in graphical user interfaces. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on GUI element localization for web agents, not manufacturing defect detection.
Multimodal Fusion of RGB and Complementary Modalities for Semantic Segmentation This paper proposes a bidirectional polarity-aware cross-modality fusion module to improve multimodal semantic segmentation across diverse complementary sensor streams. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Evaluates multi-sensor semantic segmentation on autonomous driving and general datasets, not industrial inspection.
Beyond the End-Effector: Learning Whole-Arm Contact Manipulation from Human Demonstrations This paper proposes a hierarchical framework for learning whole-arm robot contact manipulation tasks from human demonstrations. Preprint	Small and efficient models via arxiv · 15 Oct, 06:18	Not relevant Focuses on contact trajectory optimization, not model compression or efficient serving.
Beyond the End-Effector: Learning Whole-Arm Contact Manipulation from Human Demonstrations This paper proposes a hierarchical framework for learning whole-arm robot contact manipulation tasks from human demonstrations. Preprint	Computer vision for industrial applications via arxiv · 15 Oct, 06:18	Not relevant Focuses on general robotic whole-arm contact manipulation rather than visual defect inspection.
Beyond the End-Effector: Learning Whole-Arm Contact Manipulation from Human Demonstrations This paper proposes a hierarchical framework for learning whole-arm robot contact manipulation tasks from human demonstrations. Preprint	AI safety and reliability in deployment via arxiv · 15 Oct, 06:18	Not relevant Addresses physical manipulator safety constraints rather than AI deployment safety, jailbreaks, or hallucinations.

6,541–6,600 of 13,754← Previous Next →