cs.CV（2026-10-13）

📊 共 49 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (26 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (12 🔗3) 支柱三：空间感知与语义 (Perception & Semantics) (5 🔗1) 支柱一：机器人控制 (Robot Control) (3) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (26 篇)

#题目一句话要点标签🔗⭐
1 Bifocal: Training-Free Mixed Granularity Visual Token Pruning for Multimodal LLMs 提出Bifocal以解决多模态大语言模型的视觉标记修剪问题 large language model multimodal visual grounding ✅
2 ReDiS: Reasoning-Diagnosis Self-Evolution for Multimodal Reasoning in Remote Sensing 提出ReDiS框架以解决遥感多模态推理中的可靠性问题 large language model multimodal visual grounding
3 Visual Tokens on Demand: Visual Budgeting for Scalable and Efficient Multi-Image Reasoning in Multimodal LLMs 提出mmVB以解决多图像推理中的视觉预算问题 large language model multimodal
4 LLM2R: From Large Language Models to 3D Reassemblers 提出LLM2R以解决3D部件组装问题 large language model
5 Context-aware Bayesian Fusion for Multimodal Object Detection 提出上下文感知贝叶斯融合以解决多模态物体检测问题 multimodal
6 Rethinking Projector Training in Multimodal LLMs 提出PORTAL以简化多模态LLM的投影器训练 multimodal
7 Parameter-Efficient Multimodal Brain Tumor Segmentation With Stage-Wise Shared Adapters 提出阶段共享适配器以提高多模态脑肿瘤分割性能 multimodal
8 Beyond the Encoder: Object-Centric Representations Emerging from Segmentation Foundation Models 提出基于分割基础模型的对象中心表示以提升肿瘤预测 foundation model
9 Heading Hierarchy Reconstruction for Complex and Long Documents with Level-Wise Multimodal Memory 提出基于多模态记忆的框架以重构复杂文档的层级结构 multimodal
10 PolUR: Polarization Aware Multimodal Degradation Perception for Underwater Image Restoration 提出PolUR以解决水下图像恢复中的多模态退化感知问题 multimodal
11 dMRI-F: A Foundation Model for Generalizable Diffusion MRI Analysis from Sparse Acquisitions 提出dMRI-F以解决稀疏采集下的扩散MRI分析问题 foundation model
12 MicroCLIP: Connecting Text and Microscopy Images via Contrastive Language--Image Pretraining 提出MicroCLIP以解决显微镜图像与文本关联问题 large language model foundation model multimodal
13 Multimodal Fusion of RGB and Complementary Modalities for Semantic Segmentation 提出双向极性感知跨模态融合模块以解决多模态语义分割问题 multimodal
14 Benchmarking Self-Supervised 3D Medical Vision Foundation Models and Transferability Estimation 提出统一基准以评估自监督3D医学视觉基础模型的迁移能力 foundation model
15 OmniCompass: A Holistic Benchmark for Omni-Modal Video Grounding and Understanding 提出OmniCompass以解决多模态视频理解评估不足问题 large language model multimodal
16 GeoReader: Learning Latent Geometry for Spatial Reasoning in Vision-Language Models 提出GeoReader以解决视觉语言模型中的空间推理问题 foundation model chain-of-thought
17 VideoJev: System One Model for Video Understanding and Interaction 提出VideoJev以解决视频理解与交互问题 large language model multimodal
18 EmoGenie: Benchmarking and Evaluating Expressed and Evoked Emotion in T2I Generation with MLLMs 提出EmoGenie基准以评估T2I生成图像中的情感表达与唤起 large language model multimodal
19 Hallucination-aware Faithful Reasoning in Remote Sensing 提出FaithRS框架以解决遥感推理中的幻觉问题 large language model multimodal
20 Omni-RS: Unifying Multi-Task Understanding and Generation in Remote Sensing 提出Omni-RS以解决遥感领域理解与生成能力不足的问题 large language model multimodal
21 When, Where, and Why: Hierarchical Causal Explainability for Video Face Understanding with MLLMs 提出VideoFace-X-LLM框架以解决视频面部理解的可解释性问题 large language model multimodal
22 Plan-Bench: Can Existing VLMs Go Well Beyond Perception to Reason and Plan in Embodied Worlds? 提出Plan-Bench以评估VLM在具身世界中的推理与规划能力 embodied AI ✅
23 AIRRI: Towards a Human-Readable, Machine-Resistant Reading Interface 提出AIRRI以解决教育评估中的机器文本提取问题 large language model
24 RSIgen: Reciprocal Self-Improvement for Agentic Visual Generation 提出RSIgen框架以解决多模态生成中的学习分配问题 multimodal
25 MajorTOM:Elliot Pretrain, A Multi-Sensor, Multi-Temporal Pretraining Dataset for Earth Observation 提出MajorTOM:Elliot Pretrain以解决地球观测数据预训练问题 foundation model
26 MemAct: Streaming Spatial Understanding with Memory and Active Exploration 提出MemAct框架以解决流式空间理解中的记忆与主动探索问题 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (12 篇)

#题目一句话要点标签🔗⭐
27 PUGS: Physics-Guided Uncertainty Modeling for Underwater Dynamic 3D Gaussian Splatting 提出PUGS以解决水下动态场景重建中的不确定性问题 distillation 3D gaussian splatting 3DGS
28 4LC: 4D Latent Cache for Efficient Conditioning in Dynamic World Modeling 提出4LC以解决动态世界建模中的视频生成效率问题 world model world models JEPA
29 LensErase: A Foundation Model for Erase Anything 提出LensErase以解决多场景物体擦除问题 reinforcement learning foundation model
30 MERIT: Multimodal Evidence Reasoning over Image Tables 提出MERIT框架以解决图像表格推理问题 reinforcement learning multimodal
31 What Makes a Good Visual Representation for Spatial Understanding in MLLMs? 提出视觉表示方法以增强多模态大语言模型的空间理解能力 visual pre-training distillation large language model
32 DVTR-KD: Distilling Vision-Language Models via Intermediate Visual Token Reconstruction 提出DVTR-KD以解决多模态模型蒸馏效率问题 distillation large language model multimodal
33 DriveWLA: Unifying Future Dynamics and Scene Semantics through World-Language-Action Modeling for Autonomous Driving 提出DriveWLA以统一未来动态与场景语义，解决自主驾驶轨迹生成问题 reinforcement learning imitation learning world model
34 GAD: Geometry Attention Distillation for Consistent Video Generation 提出几何注意力蒸馏以解决视频生成中的一致性问题 distillation ✅
35 LongForcing: Scaling Temporal Manifold Matching for Long-Horizon World Rollout 提出LongForcing以解决长时间世界模型生成中的监督缺失问题 world model world models distillation ✅
36 Open-o3-Video-v2: Seek, Ground, and Reason with Verifiable Spatio-Temporal Evidence 提出Open-o3-Video-v2以解决视频推理中的证据获取问题 reinforcement learning TAMP ✅
37 AffectWorld: Benchmarking Emotion-Driven Future World Prediction 提出AffectWorld基准以解决情感驱动的未来世界预测问题 world model world models
38 VARI: Robust Text-to-Video Generation via Group-Structured Intent Alignment 提出VIRI以解决文本到视频生成中的语义一致性问题 DPO distillation

🔬 支柱三：空间感知与语义 (Perception & Semantics) (5 篇)

#题目一句话要点标签🔗⭐
39 Gaussian Splatting for Real-Time Scene Reconstruction: Methods, Systems, and Trade-offs 提出高效的3D高斯点云重建方法以解决实时场景重建问题 3D gaussian splatting 3DGS gaussian splatting
40 PLEDGE: Progressive Lossy Error-Bounded Compression of 3D Gaussian Splatting 提出PLEDGE以解决3D高斯点云压缩中的失真控制问题 3D gaussian splatting 3DGS gaussian splatting
41 CityCode: Building Agent-Operable 3D Cities from Sparse Aerial Observations 提出CityCode以解决稀疏航空观测下的3D城市构建问题 depth estimation metric depth
42 DT-LoRA: Reaching Beyond the Tangent Space of Low-Rank Adaptation 提出DT-LoRA以解决低秩适应性能不足问题 open-vocabulary open vocabulary ✅
43 Smartphone-Based, Field-Deployed AI Inspection System for Power Distribution Infrastructure 提出基于智能手机的AI检测系统以解决电力分配基础设施检查问题 depth estimation monocular depth

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
44 Physical-Field-Bridged Material Parameter Inference from Multi-View Deformation Videos 提出物理场桥接材料参数推断以解决变形物体操控问题 manipulation 3DGS scene reconstruction
45 QK-Edit: Revisiting Attention-based Injection in MM-DiT for Image and Video Editing 提出QK-Edit以解决MM-DiT在图像和视频编辑中的注意力注入问题 manipulation multimodal
46 FlowSim: Rendering-Free Closed-Loop Simulation for Evaluating End-to-End Driving Systems 提出FlowSim以解决闭环驾驶系统评估中的渲染问题 sim2real

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
47 Benchmarking Multimodal Large Language Models for Zero-Shot Human-Object Interaction 提出多模态大语言模型以解决零-shot人机交互问题 human-object interaction HOI large language model

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
48 EgoOmniWorld: Unifying Egocentric Understanding via Ego-Action-World Interactions 提出EgoOmniWorld以统一自我中心视频理解 egocentric large language model multimodal

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
49 Dual-Domain Sparse-View 4D CBCT Reconstruction with Spatial-Temporal Consistency Learning 提出STCNet以解决低剂量4D CBCT重建质量不足问题 geometric consistency

⬅️ 返回 cs.CV 首页 · 🏠 返回主页