1310 lines
28 KiB
JSON
1310 lines
28 KiB
JSON
[
|
|
{
|
|
"id": "2610.30108v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Multimodal Embodied Agents: A Unified Capability-centric Perspective from Computer-Use to Robot-Use",
|
|
"bytes": 1096228,
|
|
"tables": 4,
|
|
"figures": 12,
|
|
"imgs": 12,
|
|
"svgs": 12
|
|
},
|
|
{
|
|
"id": "2610.30089v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "ProactEgoMem: Benchmarking Proactive Memory in Streaming Egocentric Video",
|
|
"bytes": 1877011,
|
|
"tables": 12,
|
|
"figures": 31,
|
|
"imgs": 18,
|
|
"svgs": 710
|
|
},
|
|
{
|
|
"id": "2610.29913v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Asking a Multimodal LLM Whether Someone Fell:Zero-Shot Screening of Low-Resolution ThermalVideo in a Laboratory and an Occupied Home",
|
|
"bytes": 238446,
|
|
"tables": 6,
|
|
"figures": 12,
|
|
"imgs": 3,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.29768v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "UniRAP: Towards Unified Part-level Physical Affordance Reasoning and Actionable Perception",
|
|
"bytes": 692795,
|
|
"tables": 26,
|
|
"figures": 25,
|
|
"imgs": 15,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.29740v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "APED-CXR: Adaptive Privileged Evidence Distillation for Chest X-Ray Grounding with Image-Only Inference",
|
|
"bytes": 161546,
|
|
"tables": 7,
|
|
"figures": 5,
|
|
"imgs": 3,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.29642v1",
|
|
"http": 404,
|
|
"html": false
|
|
},
|
|
{
|
|
"id": "2610.29552v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Generalized CycleGRPO for Fine-GrainedMultimodal Understanding and Grounding",
|
|
"bytes": 644078,
|
|
"tables": 33,
|
|
"figures": 30,
|
|
"imgs": 14,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.29495v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Balancing Foundation Models and Trait-Specific Learning for Forehead-Crease Recognition",
|
|
"bytes": 192282,
|
|
"tables": 6,
|
|
"figures": 5,
|
|
"imgs": 4,
|
|
"svgs": 12
|
|
},
|
|
{
|
|
"id": "2610.29409v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "LLMSeg: Training-Free Text-Prompted Image Segmentation with B-Splines",
|
|
"bytes": 1360596,
|
|
"tables": 30,
|
|
"figures": 47,
|
|
"imgs": 343,
|
|
"svgs": 14
|
|
},
|
|
{
|
|
"id": "2610.29231v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "DisasterRoute: Evidence-Guided Dual-System Routing for Visual Disaster Question Answering",
|
|
"bytes": 212732,
|
|
"tables": 20,
|
|
"figures": 8,
|
|
"imgs": 3,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.29174v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "LocalAgent: Collaborative Agentic Verification for Fine-Grained Instance-Level Consistency",
|
|
"bytes": 312167,
|
|
"tables": 14,
|
|
"figures": 18,
|
|
"imgs": 6,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28964v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Can Quantum Layers offer Performance and Security Advantages in Vision-Language Models?",
|
|
"bytes": 229553,
|
|
"tables": 14,
|
|
"figures": 10,
|
|
"imgs": 3,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28801v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "JUICE: Unified Multimodal Modelingin Pretrained Representation Spaces",
|
|
"bytes": 1108248,
|
|
"tables": 60,
|
|
"figures": 69,
|
|
"imgs": 71,
|
|
"svgs": 17
|
|
},
|
|
{
|
|
"id": "2610.28416v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Code2Video-Compass: A Comprehensive Benchmark for General-Purpose Code-Based Video Generation across T2V, I2V, and V2V",
|
|
"bytes": 357358,
|
|
"tables": 5,
|
|
"figures": 12,
|
|
"imgs": 53,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28402v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "StreamEval: Evaluating InteractiveStreaming Video Generation Models",
|
|
"bytes": 391960,
|
|
"tables": 9,
|
|
"figures": 11,
|
|
"imgs": 9,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28382v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "RACE: Learning Multimodal Searchthrough Referential Abstractionand Compositional Evidence",
|
|
"bytes": 297518,
|
|
"tables": 19,
|
|
"figures": 18,
|
|
"imgs": 7,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28356v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "CountVideoBench: A Benchmark to Evaluate Prediction Biases on the Object Counting task by Video-Language Models",
|
|
"bytes": 350752,
|
|
"tables": 20,
|
|
"figures": 23,
|
|
"imgs": 9,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28282v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "ChartEvo: Self-Evolving Agent Harnesses for Long-Tail Chart Reasoning",
|
|
"bytes": 206359,
|
|
"tables": 9,
|
|
"figures": 17,
|
|
"imgs": 9,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28266v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "VGL-stGen: Learning a shared multimodal model for lung histology and spatial transcriptomics",
|
|
"bytes": 498064,
|
|
"tables": 16,
|
|
"figures": 34,
|
|
"imgs": 9,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28264v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "PI-MLLM: Psychophysics-Inspired Multimodal Large Language Model for Oddly Satisfying Video Emotion Analysis",
|
|
"bytes": 226615,
|
|
"tables": 23,
|
|
"figures": 13,
|
|
"imgs": 9,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28129v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "GAMF-Net: Grouped Availability-Aware Multimodal Fusion for Smartphone Skin Lesion Classification",
|
|
"bytes": 164905,
|
|
"tables": 18,
|
|
"figures": 8,
|
|
"imgs": 8,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.28087v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "WildAIGC-Bench: Can AI Image Detectors Handle the Wild? A Benchmark of Generated and Edited Images in Online Circulation",
|
|
"bytes": 279679,
|
|
"tables": 6,
|
|
"figures": 11,
|
|
"imgs": 17,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27999v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Matrix-Game 4.0: Harnessing Video Models for Real-Time Playable Game Worlds",
|
|
"bytes": 222776,
|
|
"tables": 4,
|
|
"figures": 11,
|
|
"imgs": 10,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27974v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Media Intelligence Platform: Multimodal Detection of Synthetic and Manipulated Content",
|
|
"bytes": 268375,
|
|
"tables": 5,
|
|
"figures": 12,
|
|
"imgs": 12,
|
|
"svgs": 20
|
|
},
|
|
{
|
|
"id": "2610.27904v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Staged Continual Adaptation of Multimodal Foundation Modelsfor Japanese Financial Documents",
|
|
"bytes": 439671,
|
|
"tables": 9,
|
|
"figures": 23,
|
|
"imgs": 8,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27894v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Testing Associative Creativity in Humans and Multimodal LLMswith Difficult-to-Name Visual Stimuli",
|
|
"bytes": 661015,
|
|
"tables": 24,
|
|
"figures": 40,
|
|
"imgs": 27,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27829v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "GARRET: Generative Feedback for Region Retrievalin Visual Documents",
|
|
"bytes": 429228,
|
|
"tables": 45,
|
|
"figures": 25,
|
|
"imgs": 11,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27782v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "RoboJudge: Multimodal Language Models asJudges for Embodied Video Generation",
|
|
"bytes": 437259,
|
|
"tables": 41,
|
|
"figures": 26,
|
|
"imgs": 15,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27692v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "A Principled Optimal Transport Framework for Frame Selection in Long Video Understanding",
|
|
"bytes": 475555,
|
|
"tables": 69,
|
|
"figures": 11,
|
|
"imgs": 4,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27657v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "PLaCA: Learning Patch-Level Future Latent Change for Action Generation",
|
|
"bytes": 223536,
|
|
"tables": 16,
|
|
"figures": 7,
|
|
"imgs": 8,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27571v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Text-Anchored Modality Experts with Uncertainty-Guided Routing for Multimodal Sentiment Analysis",
|
|
"bytes": 470485,
|
|
"tables": 17,
|
|
"figures": 6,
|
|
"imgs": 3,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27565v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "AnomalyJev: Towards Evaluating Image Anomaly Explanations via Claim-Level Verification",
|
|
"bytes": 253045,
|
|
"tables": 7,
|
|
"figures": 16,
|
|
"imgs": 8,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27521v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Multiscale Temporal\u2013Spectral\u2013Spatial EEG Learning With Modality-Specific and Modality-Invariant Representations for Emotion Recognition",
|
|
"bytes": 231484,
|
|
"tables": 22,
|
|
"figures": 12,
|
|
"imgs": 10,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27360v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "On the Limits of Scale: Inverse Scaling in Vision-Language Models",
|
|
"bytes": 873424,
|
|
"tables": 49,
|
|
"figures": 48,
|
|
"imgs": 24,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27341v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Look Before You Reason: Implicit Visual Thinking for Efficient Multimodal Reasoning",
|
|
"bytes": 324194,
|
|
"tables": 20,
|
|
"figures": 17,
|
|
"imgs": 10,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.27298v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Generalizing Multimodal Foundational Models in Context-Driven Class Incremental Learning",
|
|
"bytes": 523421,
|
|
"tables": 24,
|
|
"figures": 21,
|
|
"imgs": 7,
|
|
"svgs": 15
|
|
},
|
|
{
|
|
"id": "2610.27192v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "CARLAR Influence: A Structured Hallucination Risk in Multimodal Cheque Extraction and an Exploratory Compressed-Image Countermeasure",
|
|
"bytes": 228308,
|
|
"tables": 9,
|
|
"figures": 15,
|
|
"imgs": 13,
|
|
"svgs": 12
|
|
},
|
|
{
|
|
"id": "2610.26863v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Text-to-Image Generative ModelingWithout Language Pretraining",
|
|
"bytes": 637813,
|
|
"tables": 13,
|
|
"figures": 36,
|
|
"imgs": 12,
|
|
"svgs": 25
|
|
},
|
|
{
|
|
"id": "2610.26860v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory",
|
|
"bytes": 2919802,
|
|
"tables": 48,
|
|
"figures": 39,
|
|
"imgs": 131,
|
|
"svgs": 738
|
|
},
|
|
{
|
|
"id": "2610.26856v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Reinforcement Learning with Visually Attributable On-Policy Self-Distillation",
|
|
"bytes": 360721,
|
|
"tables": 19,
|
|
"figures": 20,
|
|
"imgs": 15,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.27049v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "VitronEdit: Universal Visual Editingacross Images, Videos, and 3D Scenes",
|
|
"bytes": 410485,
|
|
"tables": 28,
|
|
"figures": 18,
|
|
"imgs": 13,
|
|
"svgs": 13
|
|
},
|
|
{
|
|
"id": "2610.26855v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Ropedia W0",
|
|
"bytes": 558440,
|
|
"tables": 47,
|
|
"figures": 43,
|
|
"imgs": 32,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.26826v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "RefCraft: Benchmarking Visual-Reference-Conditioned Agentic Generation through Web Development",
|
|
"bytes": 441335,
|
|
"tables": 11,
|
|
"figures": 22,
|
|
"imgs": 11,
|
|
"svgs": 15
|
|
},
|
|
{
|
|
"id": "2610.26751v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Beyond Omni: A Unified MLLM for Embodied Sensory Understanding",
|
|
"bytes": 654761,
|
|
"tables": 34,
|
|
"figures": 36,
|
|
"imgs": 37,
|
|
"svgs": 26
|
|
},
|
|
{
|
|
"id": "2610.26585v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Drive2Gauss: FeedForward 4D Gaussian Reconstruction from Multimodal Generative Driving Videos",
|
|
"bytes": 244029,
|
|
"tables": 17,
|
|
"figures": 13,
|
|
"imgs": 13,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.26538v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Small Average Error, Wide Individual Error: General-Purpose LLMs Estimating Energy Intake from Meal Photos in a Feeding Trial",
|
|
"bytes": 277810,
|
|
"tables": 12,
|
|
"figures": 13,
|
|
"imgs": 2,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.26431v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Long-form Video-to-Text Summarization using Recursive Language Models",
|
|
"bytes": 231061,
|
|
"tables": 9,
|
|
"figures": 12,
|
|
"imgs": 8,
|
|
"svgs": 103
|
|
},
|
|
{
|
|
"id": "2610.26357v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "HGASurv: Hierarchical Granularity-Aware Multimodal Integration for Survival Prediction",
|
|
"bytes": 252694,
|
|
"tables": 8,
|
|
"figures": 7,
|
|
"imgs": 6,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.26203v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Focused Omni-Modal GUI Agents",
|
|
"bytes": 409866,
|
|
"tables": 13,
|
|
"figures": 21,
|
|
"imgs": 16,
|
|
"svgs": 12
|
|
},
|
|
{
|
|
"id": "2610.26194v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "GeoEdit: Adversarial De-Evidencing for Silent Geoprivacy Protection via Image Editing",
|
|
"bytes": 289210,
|
|
"tables": 16,
|
|
"figures": 21,
|
|
"imgs": 19,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.26084v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Context-Aware Topological Distillation Graph Network for Asymmetric EEG\u2013Eye Movement Emotion Recognition",
|
|
"bytes": 497156,
|
|
"tables": 24,
|
|
"figures": 20,
|
|
"imgs": 13,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.26078v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "DCR-TSF: Dual-Level Compositional Reasoning with Triadic Spherical Fusion for Training-Free Zero-Shot Composed Image Retrieval",
|
|
"bytes": 419850,
|
|
"tables": 26,
|
|
"figures": 38,
|
|
"imgs": 22,
|
|
"svgs": 19
|
|
},
|
|
{
|
|
"id": "2610.26046v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "AVTime: Reinforcing Dense Audio-VisualCaptions with Bidirectional Alignment",
|
|
"bytes": 795559,
|
|
"tables": 38,
|
|
"figures": 49,
|
|
"imgs": 23,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.26022v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "FrameGRPO: Incentivizing Video Reasoning via Multi-Frame Rollouts and Temporal Credit",
|
|
"bytes": 283332,
|
|
"tables": 64,
|
|
"figures": 16,
|
|
"imgs": 10,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.25999v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "TempoMoE: Identifying and Steering Fast\u2013Slow Thinking Experts for Efficient Multimodal Reasoning",
|
|
"bytes": 229449,
|
|
"tables": 13,
|
|
"figures": 6,
|
|
"imgs": 9,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.25926v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "MIRA: A Multimodal Illustrated Deep Research Agent for Controllable Reports",
|
|
"bytes": 361871,
|
|
"tables": 16,
|
|
"figures": 17,
|
|
"imgs": 10,
|
|
"svgs": 13
|
|
},
|
|
{
|
|
"id": "2610.25649v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "InteractAgent: Agentic Human Motion Interaction with Memory-Augmented Reflection",
|
|
"bytes": 213695,
|
|
"tables": 18,
|
|
"figures": 7,
|
|
"imgs": 9,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.25617v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "CheXFuse: Guided Differential Joint Attention Fusion for Robust Multimodal Chest X-Ray Classification in Data-Scarce Medical Settings",
|
|
"bytes": 165479,
|
|
"tables": 22,
|
|
"figures": 4,
|
|
"imgs": 9,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.25570v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "DiffGround: Semantic Difference-Guided Zero-Shot Video Question Grounding",
|
|
"bytes": 221117,
|
|
"tables": 22,
|
|
"figures": 14,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.25533v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Modaliff: A Unified Token Diffusion Model for Any-to-Any Generation",
|
|
"bytes": 365745,
|
|
"tables": 23,
|
|
"figures": 44,
|
|
"imgs": 37,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.25493v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "GroundedVDR: Grounding Video-to-Web Evidence Dependencies in Video Deep Research",
|
|
"bytes": 204090,
|
|
"tables": 6,
|
|
"figures": 9,
|
|
"imgs": 9,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.25481v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Multimodal-Conditioned Latent Diffusion Models forSpatiotemporal Brain MRI Forecasting",
|
|
"bytes": 222410,
|
|
"tables": 4,
|
|
"figures": 4,
|
|
"imgs": 7,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.25470v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "A failure of Imagination: VLMs fail to use visual representations for text, but self-distillation can help",
|
|
"bytes": 268682,
|
|
"tables": 7,
|
|
"figures": 16,
|
|
"imgs": 18,
|
|
"svgs": 14
|
|
},
|
|
{
|
|
"id": "2610.25448v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "UBD-Bench: Benchmarking and Analyzing Unimodal Branch Degradation in Multimodal Perception for Autonomous Driving",
|
|
"bytes": 185532,
|
|
"tables": 12,
|
|
"figures": 8,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.25418v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "From Label Priors to Task Evidence: Efficient Long-Video Frame Selection via Bayesian GRPO",
|
|
"bytes": 397244,
|
|
"tables": 28,
|
|
"figures": 25,
|
|
"imgs": 6,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.25404v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "MT Forcing: Towards Stable and ConsistentMulti-Turn Image Editing with Autoregressive Diffusion",
|
|
"bytes": 338107,
|
|
"tables": 11,
|
|
"figures": 10,
|
|
"imgs": 9,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.25247v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Visual Scaffolding:Toward Native Multimodal Reasoning",
|
|
"bytes": 302527,
|
|
"tables": 18,
|
|
"figures": 14,
|
|
"imgs": 7,
|
|
"svgs": 15
|
|
},
|
|
{
|
|
"id": "2610.25005v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "ReTrace: Visual Backtracking for Efficient Multimodal Large Language Models",
|
|
"bytes": 672583,
|
|
"tables": 47,
|
|
"figures": 359,
|
|
"imgs": 273,
|
|
"svgs": 16
|
|
},
|
|
{
|
|
"id": "2610.24786v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Video-Language Entailment for Video Similarity Estimation",
|
|
"bytes": 301618,
|
|
"tables": 13,
|
|
"figures": 17,
|
|
"imgs": 16,
|
|
"svgs": 15
|
|
},
|
|
{
|
|
"id": "2610.24771v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Design Anything, Design at Scale",
|
|
"bytes": 522309,
|
|
"tables": 29,
|
|
"figures": 41,
|
|
"imgs": 24,
|
|
"svgs": 14
|
|
},
|
|
{
|
|
"id": "2610.24537v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Write Once, Query Many: Evolving Compact Reusable Memory for Long Videos",
|
|
"bytes": 299444,
|
|
"tables": 29,
|
|
"figures": 10,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.24452v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "HumanSkill-4D: Training-Free LanguageQuerying of Animatable Gaussian Avatarsin Space and Time",
|
|
"bytes": 477205,
|
|
"tables": 33,
|
|
"figures": 28,
|
|
"imgs": 17,
|
|
"svgs": 15
|
|
},
|
|
{
|
|
"id": "2610.24395v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "MSCR: Jointly Balancing Utilization Across Modalities and Discovering Synergistic Information",
|
|
"bytes": 461252,
|
|
"tables": 27,
|
|
"figures": 11,
|
|
"imgs": 14,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.24379v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Defending Against Dormant Poisoning Attacks Across Language and Multimodal Agents",
|
|
"bytes": 502151,
|
|
"tables": 30,
|
|
"figures": 32,
|
|
"imgs": 8,
|
|
"svgs": 17
|
|
},
|
|
{
|
|
"id": "2610.24378v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "PACE-Mem: Prefix-Aware Complementary Evidence for Multimodal Agents",
|
|
"bytes": 305366,
|
|
"tables": 12,
|
|
"figures": 13,
|
|
"imgs": 5,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.24364v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "GazeAgent: VLM-Guided Implicit Gaze Calibration from Egocentric Multimodal Behavior for AI Glasses",
|
|
"bytes": 635664,
|
|
"tables": 11,
|
|
"figures": 18,
|
|
"imgs": 25,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.24239v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Encoder-Free Multimodal LLMsLearn Implicit Encoders",
|
|
"bytes": 194188,
|
|
"tables": 3,
|
|
"figures": 15,
|
|
"imgs": 6,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.24108v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Joint, Decoupled, or Agent: When Each Multimodal Alignment Paradigm Wins",
|
|
"bytes": 1420818,
|
|
"tables": 39,
|
|
"figures": 46,
|
|
"imgs": 6,
|
|
"svgs": 48
|
|
},
|
|
{
|
|
"id": "2610.24094v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "BiNar: Bitstream Native Restorer for Bitstream-corrupted Image Restoration",
|
|
"bytes": 139511,
|
|
"tables": 12,
|
|
"figures": 5,
|
|
"imgs": 10,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.23730v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Synthetic Image Oversampling for Rare-Class Representation Learning in Deep Facial Phenotyping",
|
|
"bytes": 198466,
|
|
"tables": 9,
|
|
"figures": 9,
|
|
"imgs": 9,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.23609v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "ADA 2025: A Multimodal Datasetfor Water-Deficit Stress Detection in Sugar Beets",
|
|
"bytes": 104115,
|
|
"tables": 1,
|
|
"figures": 4,
|
|
"imgs": 9,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.23588v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Enhancing RAG-based Medical VQA withGated Multimodal Representations",
|
|
"bytes": 148072,
|
|
"tables": 6,
|
|
"figures": 4,
|
|
"imgs": 7,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.23533v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "ReVision: Reprogramming Image Generators for Dense Visual Perception",
|
|
"bytes": 367632,
|
|
"tables": 34,
|
|
"figures": 18,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.23416v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "GeoReason: Bridging Logical Reasoning and Spatial Fidelity in Remote Sensing Segmentation",
|
|
"bytes": 345501,
|
|
"tables": 18,
|
|
"figures": 23,
|
|
"imgs": 17,
|
|
"svgs": 12
|
|
},
|
|
{
|
|
"id": "2610.23356v1",
|
|
"http": 404,
|
|
"html": false
|
|
},
|
|
{
|
|
"id": "2610.23324v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "CatchUp: Depth-Aware Re-Admission of Visual Tokens for Efficient Large Vision-Language Models",
|
|
"bytes": 432890,
|
|
"tables": 45,
|
|
"figures": 18,
|
|
"imgs": 10,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.23279v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Xiaomi-Lens-0 Technical Report",
|
|
"bytes": 416785,
|
|
"tables": 24,
|
|
"figures": 24,
|
|
"imgs": 16,
|
|
"svgs": 13
|
|
},
|
|
{
|
|
"id": "2610.23214v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "MLLM Frame Subset Ensembling for Audio-Visual Video QA and MLLM-based Reranking for Ad-hoc Video Search in TRECVID 2025",
|
|
"bytes": 159122,
|
|
"tables": 7,
|
|
"figures": 13,
|
|
"imgs": 8,
|
|
"svgs": 12
|
|
},
|
|
{
|
|
"id": "2610.23144v2",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "From GAN-Based Models to GPT-5: A Taxonomy-Guided Study of Human Performance in Detecting Deepfake Face Images",
|
|
"bytes": 388259,
|
|
"tables": 5,
|
|
"figures": 12,
|
|
"imgs": 8,
|
|
"svgs": 14
|
|
},
|
|
{
|
|
"id": "2610.23125v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Reliability-Consistent Appearance Modeling for RGB\u2013Infrared Multi-Object Tracking",
|
|
"bytes": 273833,
|
|
"tables": 20,
|
|
"figures": 10,
|
|
"imgs": 10,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.23110v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Domain-specific language alignment of multimodal embeddings towards explainable predictions",
|
|
"bytes": 362969,
|
|
"tables": 20,
|
|
"figures": 28,
|
|
"imgs": 8,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.22964v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "ANCHOR: Audio-Visually Grounded Chain-of-Thought Reasoning Benchmark",
|
|
"bytes": 686899,
|
|
"tables": 9,
|
|
"figures": 54,
|
|
"imgs": 46,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22907v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "COMET: Decoupled Distillation, Routing, and Capacity Control for Task-Agnostic Continual Vision\u2013Language Learning",
|
|
"bytes": 906013,
|
|
"tables": 55,
|
|
"figures": 15,
|
|
"imgs": 8,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22845v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Variational and Decomposed method for Mitigating Bimodal Bias in Robust Visual Question Answering",
|
|
"bytes": 260643,
|
|
"tables": 17,
|
|
"figures": 14,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22842v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "VTrajEmbed: Aligning Text with Video Object Trajectories via Spatiotemporal Embeddings",
|
|
"bytes": 563100,
|
|
"tables": 49,
|
|
"figures": 4,
|
|
"imgs": 7,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22822v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "VISTA: A Multimodal Egocentric Benchmark for Goal-Oriented Perception and Assistance to Blind and Low-Vision Users",
|
|
"bytes": 815502,
|
|
"tables": 33,
|
|
"figures": 28,
|
|
"imgs": 19,
|
|
"svgs": 16
|
|
},
|
|
{
|
|
"id": "2610.22744v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "OmniXRef: A Unified Framework for Multimodal Multi-Reference Video Editing",
|
|
"bytes": 617529,
|
|
"tables": 16,
|
|
"figures": 16,
|
|
"imgs": 13,
|
|
"svgs": 21
|
|
},
|
|
{
|
|
"id": "2610.22700v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Thinking from Above:Efficient Spatial Reasoning with Bird\u2019s-Eye-View Visual Tokens",
|
|
"bytes": 413326,
|
|
"tables": 13,
|
|
"figures": 14,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22602v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Stuck on Shuffle: On Attribute\u2013ObjectBinding in Vision\u2013Language Models",
|
|
"bytes": 841679,
|
|
"tables": 26,
|
|
"figures": 29,
|
|
"imgs": 12,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22509v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "TriSAR-3D and PaintFusion: A Novel Trimodal LiDAR\u2013RGB\u2013Thermal Dataset and Method for 3D Person Detection in Search and Rescue",
|
|
"bytes": 293990,
|
|
"tables": 11,
|
|
"figures": 16,
|
|
"imgs": 12,
|
|
"svgs": 14
|
|
},
|
|
{
|
|
"id": "2610.22499v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "JAVIS: Joint Audio-Visual Interaction and Streaming for Real-Time Conversational Agents",
|
|
"bytes": 142340,
|
|
"tables": 5,
|
|
"figures": 7,
|
|
"imgs": 8,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22474v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Star Light, Star Bright: Can Multimodal Models Parse an Astronomy Visual Question Answering Dataset Right?",
|
|
"bytes": 210882,
|
|
"tables": 4,
|
|
"figures": 9,
|
|
"imgs": 4,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.22360v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Grounding Generative Verification in 3D Content",
|
|
"bytes": 661125,
|
|
"tables": 21,
|
|
"figures": 36,
|
|
"imgs": 11,
|
|
"svgs": 17
|
|
},
|
|
{
|
|
"id": "2610.24005v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "PACE: Pareto-Adaptive Compression for Efficient Native MLLMs",
|
|
"bytes": 764282,
|
|
"tables": 29,
|
|
"figures": 28,
|
|
"imgs": 7,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.22276v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": ": System One Spatial Intelligencevia Native Attention Matching",
|
|
"bytes": 796241,
|
|
"tables": 47,
|
|
"figures": 36,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22240v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "MMSearch-OPD: Easy Made Hard, Hard Made Learnable for Long-Horizon Multimodal Search Agents",
|
|
"bytes": 229175,
|
|
"tables": 11,
|
|
"figures": 6,
|
|
"imgs": 8,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22234v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "OS-Helix: Unifying GUI Interaction and Code Execution in Native Computer-Using Agents",
|
|
"bytes": 760110,
|
|
"tables": 34,
|
|
"figures": 25,
|
|
"imgs": 11,
|
|
"svgs": 22
|
|
},
|
|
{
|
|
"id": "2610.22217v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Skill-Decoupled On-Policy Expert Distillation for Unified Vision-Language Models",
|
|
"bytes": 494138,
|
|
"tables": 34,
|
|
"figures": 34,
|
|
"imgs": 16,
|
|
"svgs": 18
|
|
},
|
|
{
|
|
"id": "2610.22215v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "FutureLift: Lifting Trajectory Proposals into Bidirectional Future Graphs for Motion Forecasting",
|
|
"bytes": 359441,
|
|
"tables": 17,
|
|
"figures": 30,
|
|
"imgs": 2,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.22444v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Event-STVG: Spatio-Temporal Video Grounding for Long-Form Event Reasoning",
|
|
"bytes": 364961,
|
|
"tables": 20,
|
|
"figures": 17,
|
|
"imgs": 10,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.22009v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Visual Sanad: Auditing Vision\u2013Language Models for False Acceptance of Qur\u2019anic Citations",
|
|
"bytes": 220905,
|
|
"tables": 12,
|
|
"figures": 9,
|
|
"imgs": 10,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.24002v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "spora:A Unified Multimodal Datasetand Platform for Spatial Proteomics",
|
|
"bytes": 638262,
|
|
"tables": 26,
|
|
"figures": 37,
|
|
"imgs": 22,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.21948v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Following Delegation Traces to Study Human Oversight in Interaction with Multimodal AI",
|
|
"bytes": 164692,
|
|
"tables": 8,
|
|
"figures": 6,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.21794v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "MedPAO-VL: Interpretable and Protocol-Compliant Agent for Structured Radiology Report Generation",
|
|
"bytes": 534506,
|
|
"tables": 20,
|
|
"figures": 46,
|
|
"imgs": 21,
|
|
"svgs": 42
|
|
},
|
|
{
|
|
"id": "2610.21696v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Bridging the Markerless Tracking Validation Gap: Intraoperative X-ray Workflow Integration",
|
|
"bytes": 112178,
|
|
"tables": 4,
|
|
"figures": 5,
|
|
"imgs": 8,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.21667v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "FusionAudit: Pathway-Conditioned Robustness Auditing for Native Multimodal Models",
|
|
"bytes": 292020,
|
|
"tables": 11,
|
|
"figures": 9,
|
|
"imgs": 4,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.21588v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Hear, Localize, and Reason: Spatially Aware Scene Understanding for Audio-Visual LLMs",
|
|
"bytes": 660589,
|
|
"tables": 17,
|
|
"figures": 28,
|
|
"imgs": 18,
|
|
"svgs": 11
|
|
},
|
|
{
|
|
"id": "2610.21459v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Can Visual Generation Expand What Unified Multimodal Models Know?",
|
|
"bytes": 197283,
|
|
"tables": 12,
|
|
"figures": 14,
|
|
"imgs": 11,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.21452v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Full-Duplex Interaction with Speech, Expressions and Gestures",
|
|
"bytes": 577351,
|
|
"tables": 36,
|
|
"figures": 18,
|
|
"imgs": 10,
|
|
"svgs": 10
|
|
},
|
|
{
|
|
"id": "2610.21380v1",
|
|
"http": 200,
|
|
"html": true,
|
|
"title": "Perception or Reasoning? Diagnosing and Tackling Reinforcement Learning Bottlenecks in VLMs",
|
|
"bytes": 349115,
|
|
"tables": 28,
|
|
"figures": 17,
|
|
"imgs": 22,
|
|
"svgs": 10
|
|
}
|
|
] |