VLA / 视觉-语言-动作模型

Observe Then Act: Asynchronous Active Vision-Action Model for Robotic Manipulation

Guokang Wang, Hang Li, Shuyuan Zhang, Di Guo, Yanhong Liu, Huaping Liu

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.RO、cs.CV

Transforming Multimodal Models into Action Models for Radiotherapy

Matteo Ferrante, Alessandra Carosi, Rolando Maria D Angelillo, Nicola Toschi

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Contrastive Learning to Fine-Tune Feature Extraction Models for the Visual Cortex

Alex Mulrooney, Austin J. Brockmeier

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.CV、cs.LG

xLAM: A Family of Large Action Models to Empower AI Agent Systems

Jianguo Zhang, Tian Lan, Ming Zhu, Zuxin Liu, Thai Hoang, Shirley Kokane, Weiran Yao, Juntao Tan, Akshara Prabhakar, Haolin Chen, Zhiwei Liu, Yihao Feng, Tulika Awalgaonkar, Rithesh Murthy, Eric Hu, Zeyuan Chen, Ran Xu, Juan Carlos Niebles, Shelby Heinecke, Huan Wang, Silvio Savarese, Caiming Xiong

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Comments Technical report for the Salesforce xLAM model series

Vamos: Versatile Action Models for Video Understanding

Shijie Wang, Qi Zhao, Minh Quan Do, Nakul Agarwal, Kwonjoon Lee, Chen Sun

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.CV、cs.AI

Comments Accepted to ECCV 2024 (European Conference on Computer Vision). Code and models are released at https://brown-palm.github.io/Vamos/

Multi-level Interaction Modeling for Protein Mutational Effect Prediction

Yuanle Mo, Xin Hong, Bowen Gao, Yinjun Jia, Yanyan Lan

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Transforming a Quadruped into a Guide Robot for the Visually Impaired: Formalizing Wayfinding, Interaction Modeling, and Safety Mechanism

J. Taery Kim, Wenhao Yu, Yash Kothari, Jie Tan, Greg Turk, Sehoon Ha

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.RO、cs.AI

Comments 16 pages, 8 figures

Journal ref Proceedings of The 7th Conference on Robot Learning, PMLR 229:2288-2303, 2023

FIMP: Future Interaction Modeling for Multi-Agent Motion Prediction

Sungmin Woo, Minjung Kim, Donghyeong Kim, Sungjun Jang, Sangyoun Lee

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.RO、cs.CV

Comments Accepted by ICRA 2024

Interaction Modeling with Multiplex Attention

Fan-Yun Sun, Isaac Kauvar, Ruohan Zhang, Jiachen Li, Mykel Kochenderfer, Jiajun Wu, Nick Haber

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Comments NeurIPS 2022, project website: https://cs.stanford.edu/~sunfanyun/imma/

Build generally reusable agent-environment interaction models

Jun Jin, Hongming Zhang, Jun Luo

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Comments Accepted in Foundation Models for Decision Making Workshop at Neural Information Processing Systems, 2022. Slides: https://docs.google.com/presentation/d/1PMS2xwTcztP2pPk1bsjqkQscI39Wy5tpmoE5-_ZC7Fo/edit?usp=sharing

Predicting Protein-Ligand Binding Affinity via Joint Global-Local Interaction Modeling

Yang Zhang, Gengmo Zhou, Zhewei Wei, Hongteng Xu

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Neural-Symbolic Descriptive Action Model from Images: The Search for STRIPS

Masataro Asai

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Comments Technical Report; not going to be submitted to the conference

Action Model Acquisition using LSTM

Ankuj Arora, Humbert Fiorino, Damien Pellier, Sylvie Pesty

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Comments 13 pages

Learning Action Models: Qualitative Approach

Thomas Bolander, Nina Gierasimczuk

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Comments 18 pages, accepted for LORI-V: The Fifth International Conference on Logic, Rationality and Interaction, October 28-31, 2015, National Taiwan University, Taipei, Taiwan

Structure and Parameter Learning for Causal Independence and Causal Interaction Models

Christopher Meek, David Heckerman

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI、cs.LG

Comments Appears in Proceedings of the Thirteenth Conference on Uncertainty in Artificial Intelligence (UAI1997)

Efficient Sim-to-Real Transfer of World-Action Models from Synthetic Priors

基于合成先验的世界-动作模型的高效仿真到现实迁移

Zixing Wang, Kausik Sivakumar, Jinghuan Shang, Yafei Hu, Zhaoming Xie, Ran Gong, Xiaohan Zhang, Karl Schmeckpeper

机构 * Purdue University（普渡大学） ； Robotics and AI Institute（机器人与人工智能研究所） 纠错

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.RO

AI总结 本文研究从合成先验训练世界-动作模型并零样本部署到真实机器人操作，通过域随机化和AnyTask运动规划生成演示，首次成功实现仿真到现实的迁移。

Comments This work is accepted by CVPR'26, Embodied AI Workshop. This paper represent a part of early result of our official world-action model zero-shot sim-to-real transfer work, which will be released soon

ActionStudio: A Lightweight Framework for Data and Training of Large Action Models

Jianguo Zhang, Thai Hoang, Ming Zhu, Zuxin Liu, Shiyu Wang, Tulika Awalgaonkar, Akshara Prabhakar, Haolin Chen, Weiran Yao, Zhiwei Liu, Juntao Tan, Juan Carlos Niebles, Shelby Heinecke, Huan Wang, Silvio Savarese, Caiming Xiong

专题命中 VLA模型 ：action model(title,abstract)；分类 cs.AI

Comments 16 pages; large action models; xLAM; ActionStudio

Observations of Disrupted CME Material Falling Back Into the Low Corona

回落至低日冕的受扰日冕物质抛射（CME）物质观测

Brian E. Wood, Jason E. Kooi

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 该研究利用SOHO、STEREO-A、SDO和VLA的观测数据，通过运动学阻力模型，观测到受扰CME物质回落至低日冕的现象，发现其可能有助于日冕雨的形成。

Comments 28 pages, 15 figures, to appear in The Astrophysical Journal

Don't Drop the BATON: Long-Horizon Robot Manipulation via Agentic Subtask Exploration and Transition-aware Memory

不要放弃BATON：通过智能体子任务探索和感知转换的记忆实现长程机器人操作

Bingxin Xu, Yuzhang Shang, Emilio Ferrara

机构 * University of Southern California（南加州大学） ； University of Central Florida（中佛罗里达大学） 纠错

专题命中 VLA模型 ：VLA(abstract,abstract_cn)；vision-language-action(abstract)；分类 cs.RO、cs.CV、cs.AI

AI总结 该研究针对长程机器人操作的误差累积与子任务转换问题，提出BATON方法，通过子任务探索与感知转换记忆，在RoboMemArena基准上提升任务成功率11.6%、累计成功率14.9%。

Nascent Embedded-protostar Survey in Taurus (NEST) I: Protostellar Multiplicity

金牛座新生原恒星嵌入巡天（NEST）I：原恒星多重性

Aislinn C. Plante, John J. Tobin, Patrick D. Sheehan, Noshin Yesmin, Nicholas P. Ballering, Tyler L. Bourke, Josh Eisner, Zhi-Yun Li

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 本研究通过ALMA与VLA观测结合档案数据，统计金牛座原恒星多重性，发现其多重性显著高于猎户座、英仙座，暗示金牛座保留更多原始多重系统，且存在两种不同形成途径。

Nascent Embedded-protostar Survey in Taurus (NEST) II: Measuring Dust Mass, Disk Size, and Gas Mass

金牛座新生原恒星包层嵌入调查（NEST）II：测量尘埃质量、盘大小与气体质量

Noshin Yesmin, Patrick Sheehan, John Tobin, Aislinn Coleman-Plante, Nicholas P. Ballering, Tyler L. Bourke, Josh Eisner, Hauyu Baobab Liu, Zhi-Yun Li

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 本研究针对金牛座26个原恒星盘系统，利用ALMA与VLA观测及CO同位素体建模，测量了不同波段下的尘埃、气体质量与盘尺寸，明确了金牛座原恒星盘的相关特性及气体尘埃比范围。

Comments 31 pages, 11 Figures, Figure 10 and 11 are figure sets with 28 figures and 58 figures respectively

The Quasar Feedback Survey: Ionised Outflows in type 2 QSOs with MUSE data

类星体反馈巡天：利用MUSE数据研究2型类星体中的电离外流

M. Bianchin, C. M. Harrison, T. Costa, V. Mainieri, R. A. Riffel, G. Venturi, D. Kakkad, C. Ramos Almeida, P. H. Cezar, S. Ward, A. Girdhar, S. Molyneux, L. Ulivi, E. P. Farina, J. Mullaney, F. Arrigoni Battaia

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 本研究利用VLT/MUSE及VLA数据，研究18个高光度2型类星体的电离外流，发现电子密度与外流速度正相关，强调准确电子密度测定对提升外流参数精度的重要性，为外流模拟提供参考。

Comments 23 pages, 23 figures (16 on the appendix), submitted to A&A

VLASH: Real-Time VLAs via Future-State-Aware Asynchronous Inference

VLASH: 通过未来状态感知的异步推断实现实时VLAs

Jiaming Tang, Yufei Sun, Yilong Zhao, Shang Yang, Yujun Lin, Zhuoyang Zhang, James Hou, Yao Lu, Zhijian Liu, Song Han

机构 * MIT（麻省理工学院） ； NVIDIA（英伟达） ； Tsinghua University（清华大学） ； UC Berkeley（加州大学伯克利分校） ； UCSD（加州大学圣地亚哥分校） ； Caltech（加州理工学院） 纠错

专题命中 VLA模型 ：vision-language-action(abstract)；VLA(abstract)；action model(abstract)；分类 cs.RO、cs.AI、cs.LG

AI总结 VLASH通过未来状态感知的异步推断框架，实现高效、准确的实时VLAs控制，显著提升反应速度和任务执行精度。

Dust characterization of the HD 163296 disk with high-resolution multi-wavelength ALMA observations

利用高分辨率多波长阿塔卡马大型毫米/亚毫米波阵列观测对HD 163296星系盘进行尘埃特征分析

Kiyoaki Doi, Myriam Benisty, Akimasa Kataoka, Haochang Jiang, Francesco Zagaria, Hauyu Baobab Liu, Carsten Dominik, Ryo Tazaki, Takahiro Ueda, Tomohiro Yoshida, Takashi Tsukagoshi, Yoshihide Yamato, Ryuta Orihara

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 该研究通过对HD 163296星系盘多波长高分辨率观测做SED拟合，表征其尘埃特性。利用新的ALMA波段9观测及存档数据，结合VLA观测，探索多种尘埃模型，确定DSHARP Zubko模型为首选，分析了尘埃特性及各盘尘埃质量。

Comments first revision submitted to A&A

NVIDIA OmniDreams: Real-Time Generative World Model for Closed-Loop Autonomous Vehicle Simulation

NVIDIA OmniDreams：用于闭环自动驾驶仿真的实时生成式世界模型

NVIDIA, :, Aarti Basant, Amlan Kar, Despoina Paschalidou, Fangyin Wei, Francesco Ferroni, Guillermo Garcia Cobo, Haithem Turki, Huan Ling, Jaewoo Seo, James Lucas, Jay Zhangjie Wu, Jialiang Wang, Jonathan Lorraine, Jun Gao, Kai He, Katarina Tothova, Kevin Xie, Michał Tyszkiewicz, Qi Wu, Riccardo de Lutio, Ruilong Li, Sanja Fidler, Seung Wook Kim, Tianchang Shen, Tianshi Cao, Tobias Pfaff, William Lew, Xindi Wu, Xuanchi Ren, Yifan Lu, Yuxuan Zhang, Zan Gojcic, Zian Wang

专题命中 VLA模型 ：VLA(abstract,abstract_cn)；action model(abstract)；分类 cs.RO、cs.CV、cs.AI

AI总结 提出OmniDreams，一个基于Cosmos扩散模型训练的基础生成式世界模型，通过自回归生成动作条件视频，实现闭环仿真中复杂长尾场景的实时合成，并验证其在策略模型训练中的有效性。

Learning to Fold: prizewinning solution at LeHome Challenge 2026 (1st place online, 2nd offline)

学习折叠：LeHome Challenge 2026 获奖解决方案（在线第一名，离线第二名）

Ilia Larchenko

机构 * Independent Researcher（独立研究员） 纠错

专题命中 VLA模型 ：VLA(abstract,abstract_cn)；vision-language-action(abstract)；分类 cs.RO、cs.AI、cs.LG

AI总结 提出一种结合强化学习循环的视觉-语言-动作策略，通过自值函数实现优势估计、故障检测和候选选择，在LeHome Challenge 2026中获得在线第一名和离线第二名。

Comments Solution of the LeHome Challenge at ICRA 2026

Radio Observations of the Unusual Tidal Disruption Event AT 2022wtn: a Fast and Highly Energetic Outflow

对异常潮汐瓦解事件AT 2022wtn的射电观测：快速且高能的外流

Gavin Farley, Tanmoy Laskar, Noah Franz, Collin T. Christy, Coleman Rohde, A. J. Goodwin, Kate D. Alexander, Edo Berger, Yvette Cendes, Ryan Chornock, Tarraneh Eftekhari, Walter W. Golay, Wenbin Lu, Raffaella Margutti

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 该研究通过VLA和GMRT对潮汐瓦解事件AT 2022wtn进行多时期多频率射电观测，利用均分分析框架建模并估计物理参数，对比不同几何模型，排除相对论性喷流，探讨外流起源，发现吸积盘状态转变外流符合结果，此TDE独特且强大。

Comments 20 pages, 9 figures, 5 tables

The nature of Cloud-9: a compact core embedded in a diffuse envelope

云九的本质：嵌入弥散包层的致密核心

Ruilei Zhou, Ming Zhu, Chuan-Peng Zhang, Jinlong Xu

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 该研究利用FAST和VLA数据观测云九，发现其HI有双组分结构，核心致密静止，包层延展且有速度梯度，不支持旋转，与环境相互作用有关，排除恒星组分后表明云九是暗物质主导系统，符合剥离的重子遗迹情景。

Comments 6 pages, 5 figures, Accepted by MNRAS Letters

Unveiling the Local Environment of FRB 20220912A: Sub-arcsecond $4-26$ GHz Radio Continuum Mapping

揭示快速射电暴20220912A的本地环境：亚角秒级4 - 26 GHz射电连续谱测绘

Yash Bhusare, Yogesh Maan, Mohit Bhardwaj, Thomas C. Abbott, Yuxin Dong, Danté M. Hewitt, Afrokk Khan

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 研究利用VLA对FRB 20220912A进行高分辨率多频率连续谱研究，发现未知射电源，确定其位置、直径等特征，具有特定谱指数和恒星形成率表面密度，为重复FRB起源于年轻磁星的假设提供观测支持。

Comments 16 pages, 6 figures, 4 tables

Radio and X-ray Observations of the Transitional Supernova 2019yvr: Insights into the Progenitor Mass-Loss History

过渡型超新星2019yvr的射电和X射线观测：对前身星质量损失历史的洞察

Raphael Baer-way, Poonam Chandra, Maryam Modjaz, A. J. Nayana, Keiichi Maeda, Katie Auchettl, Maria R. Drout, Charles D. Kilpatrick, Alak K. Ray, Stuart D. Ryder

专题命中 VLA模型 ：VLA(summary_cn,abstract)

AI总结 研究超新星2019yvr前身星质量损失历史，通过射电（GMRT + VLA）和X射线（Swift + Chandra）观测，结合同步加速器自吸收模型等方法，得出质量损失率下降等结果，排除相关CSM密度跃升，为理解前身星提供新认识。

Comments 19 pages,9 Figures, submitted to ApJ