cs.CV（2026-10-30）

📊 共 15 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (6 🔗1) 支柱一：机器人控制 (Robot Control) (3 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱六：视频提取与匹配 (Video Extraction) (2 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (6 篇)

#	题目	一句话要点	标签	🔗	⭐
1	SACQ: Structure-Aware Codebook Quantization for Ultra-Low-Bit Vision-Language-Action Models	提出SACQ以解决超低比特视觉-语言-动作模型的量化问题	vision-language-action VLA large language model
2	Attention Maps Know When to Think: Training-Free Difficulty-Aware Chain-of-Thought for Reasoning Segmentation	提出困难感知链式思维以解决推理分割中的过度思考问题	large language model visual grounding chain-of-thought
3	TopoHyperClinicalGraph: A Geometry and Topology-Aware Multimodal Framework for Cancer Survival Prediction	提出TopoHyperClinicalGraph以解决癌症生存预测问题	multimodal
4	Open-Weight MLLMs Cannot Compare What They Cannot Name	探讨多模态大语言模型在生物图像分析中的局限性	large language model multimodal	✅
5	Rethinking Asymmetry in Self-Supervised 3D Geometry Learning	提出基于3D表示空间的不对称自监督学习方法以提升几何估计	foundation model
6	Parameter-Efficient Fine-Tuning of Transformer Encoders for Audio-Image-Symbolic Music Retrieval	提出参数高效微调方法以优化多模态音乐检索	multimodal

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#	题目	一句话要点	标签	🔗	⭐
7	FlowMoDE: Coarse-to-fine Flow Matching for Detail-aware Sim-to-Real Monocular Depth Estimation	提出FlowMoDE以解决单目深度估计效率与精度问题	sim-to-real flow matching depth estimation
8	MotorSense: A Video-EMG Dataset of Motoric Representations for Action Understanding	提出MotorSense数据集以解决动作理解中的信号缺失问题	manipulation bi-manual representation learning
9	SketchDirector: Complex Scene Sketching via Agentic Layout Reasoning and Detail Refinement	提出SketchDirector以解决复杂场景草图生成问题	manipulation large language model	✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#	题目	一句话要点	标签	🔗	⭐
10	Latent Reasoning in Continuous Space for Unified Multimodal Models	提出LARC框架以解决多模态模型推理与生成的矛盾问题	reinforcement learning large language model multimodal
11	GRIT: Gradient Masking for Visually Misgrounded Tokens in Text-Rich Reasoning	提出GRIT框架以解决视觉信息误解问题	reinforcement learning visual grounding
12	Bayesian PET Reconstruction with Learned Flow Matching Priors via Langevin Sampling	提出基于流匹配生成先验的贝叶斯PET重建方法以解决低剂量图像重建问题	flow matching

🔬 支柱六：视频提取与匹配 (Video Extraction) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
13	DUGO: Dual-Scale Gaussian Memory for Online Temporal Action Localization in Egocentric Procedural Videos	提出DUGO框架以解决在线时序动作定位问题	egocentric	✅
14	Hawkeye: Reading Illegible Text with Vision Language Models	提出Hawkeye以解决模糊文本识别问题	egocentric

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
15	GenDSR: Transferring Representations from Video Generation Models for Dynamic Spatial Reasoning	提出GenDSR以解决动态空间推理问题	spatiotemporal

⬅️ 返回 cs.CV 首页 · 🏠 返回主页