机器人 / 具身智能 - arXivDaily 专题

2207.11232 2023-04-11 cs.CV cs.AI cs.GR cs.LG 67%

Neural Groundplans: Persistent Neural Scene Representations from a Single Image

Prafull Sharma, Ayush Tewari, Yilun Du, Sergey Zakharov, Rares Ambrus, Adrien Gaidon, William T. Freeman, Fredo Durand, Joshua B. Tenenbaum, Vincent Sitzmann

专题命中具身推理：robotics(abstract)；分类 cs.AI、cs.CV、cs.LG

Comments Project page: https://prafullsharma.net/neural_groundplans/

2302.10320 2023-02-22 cs.LG cs.AI cs.NE cs.RO 67%

Meta-World Conditional Neural Processes

Suzan Ece Ada, Emre Ugur

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments 11 pages, 9 figures

2302.08669 2023-02-20 cs.LG cs.AI cs.RO 67%

Learning to Forecast Aleatoric and Epistemic Uncertainties over Long Horizon Trajectories

Aastha Acharya, Rebecca Russell, Nisar R. Ahmed

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments Accepted to ICRA 2023

2211.15327 2022-11-29 cs.AI cs.CV cs.RO eess.IV 67%

Task-Aware Asynchronous Multi-Task Model with Class Incremental Contrastive Learning for Surgical Scene Understanding

Lalithkumar Seenivasan, Mobarakol Islam, Mengya Xu, Chwee Ming Lim, Hongliang Ren

专题命中具身推理：robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

Comments Manuscript accepted in the International Journal of Computer Assisted Radiology and Surgery. codes available: https://github.com/lalithjets/Domain-adaptation-in-MTL

2210.03043 2022-10-07 cs.CV cs.LG cs.RO 67%

Feature-Realistic Neural Fusion for Real-Time, Open Set Scene Understanding

Kirill Mazur, Edgar Sucar, Andrew J. Davison

专题命中具身推理：robotics(abstract)；分类 cs.RO、cs.CV、cs.LG

Comments For our project page, see https://makezur.github.io/FeatureRealisticFusion/

2203.12670 2022-03-25 cs.LG cs.AI cs.HC cs.NE cs.RO 67%

Competency Assessment for Autonomous Agents using Deep Generative Models

Aastha Acharya, Rebecca Russell, Nisar R. Ahmed

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.AI、cs.LG

2111.15318 2022-01-11 cs.CV cs.GR cs.LG cs.RO 67%

DiffSDFSim: Differentiable Rigid-Body Dynamics With Implicit Shapes

Michael Strecke, Joerg Stueckler

专题命中具身推理：robotics(abstract)；分类 cs.RO、cs.CV、cs.LG

Comments 22 pages, 23 Figures (including supplementary material). Presented 3DV 2021. Project website: https://diffsdfsim.is.tue.mpg.de/

Journal ref 2021 International Conference on 3D Vision (3DV)

2104.00387 2021-04-02 cs.AI cs.CV cs.RO 67%

Commonsense Spatial Reasoning for Visually Intelligent Agents

Agnese Chiatti, Gianluca Bardaro, Enrico Motta, Enrico Daga

专题命中具身推理：robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

2103.00586 2021-03-02 cs.RO cs.AI cs.CV 67%

A Kinematic Bottleneck Approach For Pose Regression of Flexible Surgical Instruments directly from Images

Luca Sestini, Benoit Rosa, Elena De Momi, Giancarlo Ferrigno, Nicolas Padoy

专题命中具身推理：robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

2011.10452 2020-11-23 cs.RO cs.AI cs.CV 67%

Bridging Scene Understanding and Task Execution with Flexible Simulation Environments

Zachary Ravichandran, J. Daniel Griffith, Benjamin Smith, Costas Frost

专题命中具身推理：robotics(abstract)；分类 cs.RO、cs.AI、cs.CV

1912.01603 2020-03-18 cs.LG cs.AI cs.RO 67%

Dream to Control: Learning Behaviors by Latent Imagination

Danijar Hafner, Timothy Lillicrap, Jimmy Ba, Mohammad Norouzi

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.AI、cs.LG

Comments 9 pages, 12 figures

1907.03950 2019-11-26 cs.AI cs.CL cs.CV cs.LG 67%

Learning by Abstraction: The Neural State Machine

Drew A. Hudson, Christopher D. Manning

专题命中具身推理：world model(abstract)；分类 cs.AI、cs.CV、cs.LG

Comments Published as a conference paper at NeurIPS 2019 (spotlight)

1902.05485 2019-10-14 cs.NE 67%

Engineered Self-Organization for Resilient Robot Self-Assembly with Minimal Surprise

Tanja Katharina Kaiser, Heiko Hamann

专题命中具身推理：world model(abstract)；robotic(abstract)

1904.12738 2019-04-30 cs.RO cs.AI cs.CV cs.NE 67%

Self Training Autonomous Driving Agent

Shashank Kotyan, Danilo Vasconcellos Vargas, Venkanna U

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.AI、cs.CV

1904.04862 2019-04-11 cs.LG cs.AI cs.CV stat.ML 67%

SWNet: Small-World Neural Networks and Rapid Convergence

Mojan Javaheripi, Bita Darvish Rouhani, Farinaz Koushanfar

专题命中具身推理：world model(abstract)；分类 cs.AI、cs.CV、cs.LG

1805.07029 2018-05-21 cs.CV cs.AI cs.LG 67%

Scene Understanding Networks for Autonomous Driving based on Around View Monitoring System

JeongYeol Baek, Ioana Veronica Chelu, Livia Iordache, Vlad Paunescu, HyunJoo Ryu, Alexandru Ghiuta, Andrei Petreanu, YunSung Soh, Andrei Leica, ByeongMoon Jeon

专题命中具身推理：navigation(abstract)；分类 cs.AI、cs.CV、cs.LG

Comments Accepted by CVPR 2018 Workshop on Autonomous Driving

1802.07461 2018-02-22 cs.LG cs.AI cs.CV stat.ML 67%

Emergence of Structured Behaviors from Curiosity-Based Intrinsic Motivation

Nick Haber, Damian Mrowca, Li Fei-Fei, Daniel L. K. Yamins

专题命中具身推理：world model(abstract)；分类 cs.AI、cs.CV、cs.LG

Comments 6 pages, 5 figures

1608.00359 2016-08-02 cs.RO cs.AI cs.LG 67%

Discovering Latent States for Model Learning: Applying Sensorimotor Contingencies Theory and Predictive Processing to Model Context

Nikolas J. Hemion

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.AI、cs.LG

1110.5102 2011-10-25 cs.CV cs.AI cs.RO 67%

Towards Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models

Congcong Li, Adarsh Kowdle, Ashutosh Saxena, Tsuhan Chen

专题命中具身推理：robotic(abstract)；分类 cs.RO、cs.AI、cs.CV

Comments 14 pages, 11 figures

2608.09266 2026-08-11 cs.CV cs.LG 新提交 66%

Did the Grid Erase the Event? EndoClock for Auditing Medical World-Model Pipelines

网格是否抹去了事件？用于审计医疗世界模型流程的EndoClock

Yarin Udi, Tom Sharon-Shahak, Roee Masad, Dan Pri-Tal

机构 * Diastole Medical R&D（Diastole医疗研发机构）

专题命中具身推理：world model(abstract,comments)；分类 cs.CV、cs.LG

AI总结本研究针对医疗世界模型同步预处理可能抹去任务相关证据的问题，提出EndoClock审计方法及四分类法，以超声心动图为例验证其有效性，为医疗AI流程审计提供可执行方案。

Comments Accepted for publication at the 1st MICCAI Workshop on Medical World Models (MWM 2026)

2606.21456 2026-06-23 cs.CV cs.RO 新提交 66%

Technical Report for ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge: Exploring Query-Based Segmentation and Increased Spatial Context for Outdoor Scene Understanding

ICRA 2026 GOOSE 2D细粒度语义分割挑战赛技术报告：探索基于查询的分割和增加空间上下文用于户外场景理解

David Pascual-Hernández, Roberto Calvo-Palomino, Inmaculada Mora-Jiménez, Jose María Cañas-Plaza

机构 * Rey Juan Carlos University（胡安卡洛斯国王大学）

专题命中具身推理：robotics(abstract,comments)；分类 cs.RO、cs.CV

AI总结本报告提出基于SegFormer和Mask2Former的细粒度语义分割方法，通过增大训练裁剪尺寸和测试时增强，在GOOSE挑战赛上达到69.6% mIoU，验证了查询式分割和空间上下文的重要性。

Comments Ranked 5th in the GOOSE 2D Fine-Grained Semantic Segmentation Challenge at the IEEE ICRA 2026 Workshop on Field Robotics

1806.06498 2026-06-04 cs.RO cs.LG cs.SY eess.SY 66%

Conditional Affordance Learning for Driving in Urban Environments

面向城市环境的条件性 affordance 学习

Axel Sauer, Nikolay Savinov, Andreas Geiger

机构 * Chair of Robotics Science and System Intelligence, Technical University of Munich（慕尼黑技术大学机器人科学与系统智能系）

专题命中具身推理：navigation(abstract)；分类 cs.RO、cs.LG；robot learning(comments)

AI总结本文提出了一种直接感知方法，通过将视频输入映射到适合复杂城市环境自主导航的中间表示，结合高层方向输入，实现了比现有强化学习和条件模仿学习方法更高的目标导向导航性能，并首次通过图像级标签处理交通灯和速度标志，显著减少模拟中的交通事故。

Comments Accepted for Conference on Robot Learning (CoRL) 2018

2509.14967 2025-09-22 cs.RO cs.HC 65%

Affordance-Based Disambiguation of Surgical Instructions for Collaborative Robot-Assisted Surgery

Ana Davila, Jacinto Colan, Yasuhisa Hasegawa

机构 * Nagoya University, Japan（名古屋大学）

专题命中具身推理：robotic(abstract,comments)；分类 cs.RO；robot learning(comments)

Comments To be presented at the 1st Workshop on Intelligent Cobodied Assistance and Robotic Empowerment (iCARE). 2025 Conference on Robot Learning (CoRL)

2608.23486 2026-08-26 cs.CV cs.RO 版本更新 62%

GeoWAM: Visual Geometry World Action Models for Autonomous Driving

GeoWAM：面向自动驾驶的视觉几何世界动作模型

Yiren Lu, Xin Ye, Jiaming Liu, Philip Jacobson, Jin Yao, Yi-chung Chen, Liam Merino, Dhruva Dixith Kurra, Min Cai, Tom Lampo, Yu Yin, Danhua Guo, Burhan Yaman

机构 * Uber AV Labs（优步自动驾驶实验室）； Case Western Reserve University（凯斯西储大学）

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.CV

AI总结该研究提出 GeoWAM，一种用于自动驾驶的视觉几何世界动作模型，通过预训练预测未来场景几何来学习动力学，经评估其生成的驾驶策略比图像基方案更强，确立未来几何预测为自动驾驶有效预训练目标。

Comments Project page: https://yiren-lu.com/project_pages/geowam/

2608.22642 2026-08-25 cs.LG cs.AI 新提交 62%

Mol-JEPA: A multimodal Joint Embedding Predictive Architecture for Molecules

Mol-JEPA：用于分子的多模态联合嵌入预测架构

Florian Rottach, Sebastian Schieferdecker, William Rudman, Randall Balestriero, Carsten Eickhoff

机构 * University of Tübingen（蒂宾根大学）； Boehringer Ingelheim（勃林格殷格翰）； The University of Texas at Austin（德克萨斯大学奥斯汀分校）； Brown University（布朗大学）

专题命中具身推理：world model(abstract)；分类 cs.AI、cs.LG

AI总结针对分子基础模型的化学无效增强等局限，提出Mol-JEPA多模态框架，利用模态掩码融入生化上下文，在基准测试中展现出优异性能。

2608.20891 2026-08-24 cs.RO cs.CV 新提交 62%

IMU-Free Body-Frame State Estimation with Sparse Scene Flow for Quadcopters

面向四旋翼无人机的基于稀疏场景流的无IMU机体坐标系状态估计

Daniel Grønhaug, Sofie Markeset, Mathias Kolberg

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.CV

AI总结该研究提出一种无IMU的四旋翼仅视觉状态估计系统，采用扩展卡尔曼滤波与4视图光束平差法，生成机体坐标系状态估计与稀疏场景流，无需GPS等世界坐标系基础设施。

Comments 56 pages, 5 figures, 2 tables. Evaluated on the VID dataset (arXiv:2103.11152)

2606.01444 2026-08-24 cs.AI cond-mat.mtrl-sci cs.CL cs.LG math.CT 版本更新 62%

Self-Revising Discovery Systems for Science: A Categorical Framework for Agentic Artificial Intelligence

科学中的自我修正发现系统：面向主体人工智能的范畴论框架

Fiona Y. Wang, Markus J. Buehler

机构 * Laboratory for Atomistic and Molecular Mechanics（原子分子力学实验室）； Department of Biological Engineering（生物工程系）； Massachusetts Institute of Technology（麻省理工学院）； Department of Civil and Environmental Engineering（土木与环境工程系）； Department of Mechanical Engineering（机械工程系）； Center for Computational Science and Engineering（计算科学与工程中心）； Schwarzman College of Computing（施瓦茨曼计算学院）

专题命中具身推理：world model(abstract)；分类 cs.AI、cs.LG

AI总结本文提出一个基于范畴论的框架，通过左Kan扩展实现科学发现中的表征体制转换，并应用于材料科学中的蛋白质力学和纤维网络建模。

2608.19492 2026-08-21 cs.LG cs.RO 新提交 62%

Beyond Multimodal Alignment: Certifying Physical Language through Response Substitution and Ordered Execution

超越多模态对齐：通过响应替换与有序执行验证物理语言

Kaizhen Tan, Xin Xu, Siru Tao, Yixiao Li, Hanzhe Hong, Yang Feng, Heqing Du

机构 * New York University（纽约大学）； Carnegie Mellon University（卡内基梅隆大学）； Columbia University（哥伦比亚大学）

专题命中具身推理：world model(abstract)；分类 cs.RO、cs.LG

AI总结本文提出DBOSC方法，在Cluster Haptic数据集与弹塑性系统中验证了多模态物理表示的可执行含义，明确了执行器、图表等因素对动作组合的影响，区分了多项可测试的操作能力成就。

2608.14530 2026-08-17 cs.CV cs.AI 新提交 62%

Marionette: Predicting World States, Rendering Geometry, Painting Appearance

Marionette：预测世界状态、渲染几何、绘制外观

Zian Meng, Zhen Li, Chuanhao Li, Qiang Li, Kaipeng Zhang

机构 * Alaya Lab（Alaya实验室）； Shanghai Innovation Institute（上海创新研究院）； Huazhong University of Science and Technology（华中科技大学）

专题命中具身推理：world model(abstract)；分类 cs.AI、cs.CV

AI总结 Marionette是面向带关节角色的交互式游戏的世界模型，通过显式建模3D世界状态、委托几何计算、合成外观，实现了可控的长时序行为，且外观生成无明显保真度损失。

Comments Project page: https://alayalab.github.io/Marionette/

2608.13095 2026-08-14 cs.RO cs.CV 新提交 62%

Semantic Radiance Fields as Simulators for Spatial Reasoning in Real-World Scenes

语义辐射场作为真实场景空间推理的模拟器

Nico Heider, Michał Jan Włodarczyk, Katarzyna Wasielewska-Michniewska, Przemysław Hołda, Martin Schieck, Marcin Paprzycki, Maria Ganzha, Bogdan Franczyk

机构 * Leipzig University（莱比锡大学）； Systems Research Institute Polish Academy of Sciences（波兰科学院系统研究所）； Wrocław University of Economics（弗罗茨瓦夫经济大学）

专题命中具身推理：embodied agent(abstract)；分类 cs.RO、cs.CV

AI总结该研究提出用语义辐射场（SRF）构建兼具几何真实性与语义可查询性的真实场景模拟器，用于训练评估具身智能体的空间推理能力，还将其应用于果园苹果采摘任务。

Comments Accepted at the IJCAI 2026 Workshop on Spatio-Temporal Reasoning and Learning (STRL), oral presentation

1. 具身推理 3140 篇

Neural Groundplans: Persistent Neural Scene Representations from a Single Image

Meta-World Conditional Neural Processes

Learning to Forecast Aleatoric and Epistemic Uncertainties over Long Horizon Trajectories

Task-Aware Asynchronous Multi-Task Model with Class Incremental Contrastive Learning for Surgical Scene Understanding

Feature-Realistic Neural Fusion for Real-Time, Open Set Scene Understanding

Competency Assessment for Autonomous Agents using Deep Generative Models

DiffSDFSim: Differentiable Rigid-Body Dynamics With Implicit Shapes

Commonsense Spatial Reasoning for Visually Intelligent Agents

A Kinematic Bottleneck Approach For Pose Regression of Flexible Surgical Instruments directly from Images

Bridging Scene Understanding and Task Execution with Flexible Simulation Environments

Dream to Control: Learning Behaviors by Latent Imagination

Learning by Abstraction: The Neural State Machine

Engineered Self-Organization for Resilient Robot Self-Assembly with Minimal Surprise

Self Training Autonomous Driving Agent

SWNet: Small-World Neural Networks and Rapid Convergence

Scene Understanding Networks for Autonomous Driving based on Around View Monitoring System

Emergence of Structured Behaviors from Curiosity-Based Intrinsic Motivation

Discovering Latent States for Model Learning: Applying Sensorimotor Contingencies Theory and Predictive Processing to Model Context

Towards Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models

Did the Grid Erase the Event? EndoClock for Auditing Medical World-Model Pipelines

Technical Report for ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge: Exploring Query-Based Segmentation and Increased Spatial Context for Outdoor Scene Understanding

Conditional Affordance Learning for Driving in Urban Environments

Affordance-Based Disambiguation of Surgical Instructions for Collaborative Robot-Assisted Surgery

GeoWAM: Visual Geometry World Action Models for Autonomous Driving

Mol-JEPA: A multimodal Joint Embedding Predictive Architecture for Molecules

IMU-Free Body-Frame State Estimation with Sparse Scene Flow for Quadcopters

Self-Revising Discovery Systems for Science: A Categorical Framework for Agentic Artificial Intelligence

Beyond Multimodal Alignment: Certifying Physical Language through Response Substitution and Ordered Execution

Marionette: Predicting World States, Rendering Geometry, Painting Appearance

Semantic Radiance Fields as Simulators for Spatial Reasoning in Real-World Scenes