cs.CV（2026-06-16）

📊 共 42 篇论文 | 🔗 13 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (15 🔗8) 支柱二：RL算法与架构 (RL & Architecture) (13 🔗3) 支柱三：空间感知与语义 (Perception & Semantics) (8 🔗1) 支柱一：机器人控制 (Robot Control) (3 🔗1) 支柱四：生成式动作 (Generative Motion) (2) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#题目一句话要点标签🔗⭐
1 Predicting Immune Biomarkers with MultiModal Mixture-of-Expert Pathology Foundation Models Empowers Precision Oncology 提出MixTIME以解决肿瘤免疫微环境生物标志物预测问题 foundation model multimodal
2 Million-scale multimodal pollen microscopy with expert-guided foundation models 提出百万规模多模态花粉显微镜资源以解决自动化识别瓶颈问题 foundation model multimodal
3 Seeing Is Not Screening: Multimodal Hidden Instruction Attacks on Agent Skill Scanners 提出SkillCamo与ExecScan以解决多模态隐藏指令攻击问题 multimodal
4 MLLMs Get It Right, Then Get It Wrong: Tracing and Correcting Late-Layer Textual Bias 提出CALRD方法以解决多模态大语言模型的文本偏见问题 large language model multimodal visual grounding
5 Plug-and-Adapt: Multimodal Coreference Resolution at First Sight with a Pretrained Alignment Model 提出Plug-and-Adapt方法以解决多模态共指消解问题 multimodal
6 A Quantitative Analysis of Multimodal Biomarkers in Alzheimer's Disease 提出多模态生物标志物定量分析以改善阿尔茨海默病研究 multimodal ✅
7 SegTME-UNI2: A Foundation Model-Based Framework for Generalisable Multiclass Cell Segmentation and LLM-Driven Tumour Microenvironment Characterisation in Histopathology 提出SEGTME-UNI2框架以解决肿瘤微环境表征问题 foundation model
8 Universal Image Restoration via Internalized Chain-of-Thought Reasoning 提出CoTIR以解决复杂图像恢复问题 chain-of-thought ✅
9 UoU: A Universal Fingerprint Foundation Model Based on Large-Scale Unsupervised Learning 提出UoU模型以解决指纹识别领域的特定任务限制问题 foundation model ✅
10 Visuals Lie, Consistency Speaks: Disentangling Spatial Attention from Reliability in Vision-Language Models 提出VLM可靠性探测器以解决多模态模型的信任问题 foundation model multimodal visual grounding
11 Beyond Visual Cues: CoT-Enhanced Reasoning for Semi-supervised Medical Image Segmentation 提出CERS框架以解决医学图像分割中的视觉语义不匹配问题 large language model chain-of-thought ✅
12 LADBench: A Benchmark for Logical Fault Detection in Images 提出LADBench以解决视觉模型的逻辑故障检测问题 foundation model multimodal ✅
13 PhaseWin: An Efficient Search Algorithm for Faithful Visual Attribution 提出PhaseWin以解决视觉归因中的高计算成本问题 visual grounding ✅
14 LiveStarPro: Proactive Streaming Video Understanding with Hierarchical Memory for Long-Horizon Streams 提出LiveStarPro以解决长视频流的主动理解问题 large language model ✅
15 StereoFactory: A Unified Merging Framework for Robust Stereo Matching 提出StereoFactory以解决立体匹配模型合并问题 foundation model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (13 篇)

#题目一句话要点标签🔗⭐
16 Future Dynamic 3D Reconstruction: A 3D World Model with Disentangled Ego-Motion 提出FR3D以解决动态环境未来重建问题 world model world models teacher-student
17 MambaCount: Efficient Text-guided Open-vocabulary Object Counting with Spatial Sparse State Space Duality Block 提出MambaCount以解决文本引导的开放词汇物体计数问题 Mamba MAE open-vocabulary
18 EgoCS-400K: An Egocentric Gameplay Dataset for World Models 提出EgoCS-400K以解决大规模互动世界建模数据不足问题 world model world models egocentric
19 ActWorld: From Explorable to Interactive World Model via Action-Aware Memory 提出ActWorld以解决交互世界模型的导航与对象交互缺失问题 world model world models human-object interaction ✅
20 See First, Answer Later: Visual Evidence Pre-Alignment via Sufficiency-Driven RL 提出视觉证据预对齐方法以解决多模态模型的视觉推理问题 reinforcement learning large language model multimodal
21 RT-Counter: Real-Time Text-Guided Open-Vocabulary Object Counting 提出RT-Counter以解决实时文本引导的开放词汇物体计数问题 MAE open-vocabulary open vocabulary ✅
22 Unified Multimodal Autoregressive Modeling with Shared Context-Visual Tokenizer is Key to Unification 提出UniAR以解决多模态建模中视觉表示分离问题 reinforcement learning multimodal ✅
23 MaineCoon: Pursuing A Real-Time Audio-Visual Social World Model 提出MaineCoon以解决社交世界模型生成的实时性问题 world model world models distillation
24 OmniDrive: An LLM-Choreographed Multi-Agent World Model with Unified Latent Co-Compression for Multi-View Driving Video Generation 提出DRIVE-CHOREO以解决多视角驾驶视频生成中的控制注入与几何融合问题 world model world models
25 BrainWorld: A Structural-Prior-Conditioned Generative Model for Whole-Brain 4D fMRI Dynamics 提出BrainWorld以解决全脑4D fMRI生成问题 representation learning foundation model multimodal
26 Reload-Mamba: Hierarchical Anti-Dilution State-Space Modeling for Multi-Class Semantic Segmentation 提出Reload-Mamba以解决多类语义分割中的响应稀释问题 Mamba state space model
27 Reinforcing Dual-Path Reasoning in Spatial Vision Language Models 提出SR-REAL以解决复杂空间推理问题 reinforcement learning chain-of-thought
28 Theoretical Grounding of Out-Of-Distribution Detection With Reinforcement Learning Optimizer 提出基于强化学习优化器的动态OOD检测方法 reinforcement learning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (8 篇)

#题目一句话要点标签🔗⭐
29 MoonSplat: Monocular Online Gaussian Splatting with Sim(3) Global Optimization 提出MoonSplat以解决单目在线3D重建中的相机姿态估计问题 3D gaussian splatting 3DGS 3D reconstruction ✅
30 AIGS-Net: Compact Illumination Field Modeling via 2D Gaussian Splatting for Fast Low-Light Image Enhancement 提出AIGS-Net以解决低光照图像增强中的计算复杂性问题 gaussian splatting splatting
31 Test-Time Training for Robust Text-Guided Open-Vocabulary Object Counting 提出Dual-TTT以解决TOOC在恶劣环境下的鲁棒性问题 open-vocabulary open vocabulary
32 Gaussian Light Field Splatting: A Physical Prior-Driven Vision Transformer for Unsupervised Low-Light Image Enhancement 提出GLFS以解决低光照图像增强中的曝光不均与色彩失真问题 gaussian splatting splatting
33 TerraTransfer: Learning End-to-End Driving Policies Without Expert Demonstrations 提出TerraTransfer以解决无专家示范的端到端驾驶策略学习问题 3D gaussian splatting gaussian splatting splatting
34 Neural Tree Reconstruction for the Open Forest Observatory 提出神经树木重建方法以解决森林映射精度不足问题 3D reconstruction NeRF neural radiance field
35 Improving and Evaluating Hand-Object Interaction Detection 提出HOI-DETR框架以提升手-物交互检测性能 3D reconstruction HOI
36 GSPan: A Continuous Gaussian Primitive Representation for Arbitrary-Scale Pansharpening 提出GSPan以解决任意尺度的全色图像融合问题 gaussian splatting splatting

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
37 WeaveLA: Event Driven Cross-Subtask Latent Memory Weaving for Repetitive Robot Manipulation 提出WeaveLA以解决跨子任务信息传递问题 manipulation vision-language-action VLA
38 GeneralVLA-2: Geometry-Aware Reconstruction and Governed Memory for Robot Planning 提出GeoFuse-MV3D与知识库升级以解决机器人规划中的重建与记忆问题 manipulation 3D reconstruction sam 3D ✅
39 Contact-Based Fringe Projection Profilometry for High-Resolution 3-D Surface Measurement of Reflective and Transparent Objects 提出基于接触的条纹投影技术以解决高反射和透明物体的3D测量问题 manipulation

🔬 支柱四：生成式动作 (Generative Motion) (2 篇)

#题目一句话要点标签🔗⭐
40 Geometric Consistency Protocol for Foundation Model Features in Multi-View Satellite Imagery 提出几何一致性协议以解决多视角卫星影像特征评估问题 physically plausible geometric consistency foundation model
41 MOCHI: Motion Enhancement of Collaborative Human-object Interactions 提出MOCHI以解决协作人机交互数据噪声问题 physically plausible human-object interaction

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

#题目一句话要点标签🔗⭐
42 Divide, Deliberate, Decide: A Multi-Agent Framework for Fine-Grained Egocentric Action Recognition 提出多代理框架以解决细粒度自我中心动作识别问题 egocentric

⬅️ 返回 cs.CV 首页 · 🏠 返回主页