MP
Marco Pavone
cs.ROcs.SYeess.SYcs.LGcs.CVcs.AImath.OCcs.MAcs.CLcs.GT
On Valency
published · living versionsW_z6b3gjgg·v1 · currentpublished
Should the Action Expert Train the VLM? Action-Token Supervision and Gradient Routing in Mixture-of-Transformer VLAs
with Asher James Hancock, Boyi Li, Yuxiao Chen, Shuhan Tan +4
1 version
W_ybywrwg8·v1published
Preprints & journals
397 papers in the corpus · 2009–2026GGraph: A GPU-Accelerated Library for Graph-Based Robot Deployment and Learning2610.28953v1 · Yash Rampuria, Rohan Garg, Daniele Gammelli et al.2026 · 0 citationsarXivon Valency
Full-Order Safety for Nonlinear Reduced-Order MPC2610.19250v1 · Luis Pabon, Johannes Kohler, Revant Kasichainula et al.2026 · 0 citationsarXiv
BT-Agent: Deploying Onboard Robot Agents with Behavior Trees and Multi-Fidelity Evaluation2610.13862v1 · Fabio Hubel, Adam Lilja, Yue Yin et al.2026 · 0 citationsarXiv
Parametric Trajectory Distillation for Few-Step Video Generation2610.11498v1 · Lan Feng, Peter Karkus, Maximilian Igl et al.2026 · 0 citationsarXiv
PLUTO: An Agentic AI Tool for Interactive Spacecraft Rendezvous Trajectory Design2610.07573v3 · Eleanor Brosius, Yuji Takubo, Daniele Gammelli et al.2026 · 0 citationsarXiv
VeriFine: Scaling Verification for Self-Improvement in Embodied Reasoning2610.08761v1 · Zewei Zhou, Rachel Luo, Yulong Cao et al.2026 · 0 citationsarXiv
Careful Judge: Safe and Efficient Human-AI Collaborative Decision Making2610.09043v1 · Chenyu Zhang, Rachel Luo, Boyi Li et al.2026 · 0 citationsarXiv
Co-Evolving Robot Orchestrators and Policies through Deployment2610.09228v1 · Xilun Zhang, Maggie Wang, Erik Bauer et al.2026 · 0 citationsarXiv
Grounding What Shapes the Plan: Rethinking Groundedness for Physical Intelligence in Autonomous Driving2610.07521v1 · Minkyoung Cho, Zewei Zhou, Wenhao Ding et al.2026 · 0 citationsarXiv
CtrlWAM: Controllable World Action Models with Aligned Intent and Foresight2610.00859v1 · Chensheng Peng, Wenhao Ding, Ran Tian et al.2026 · 0 citationsarXiv
OrbitTAMP: Grounding Language Models for Task and Motion Planning in Spacecraft Rendezvous2610.01093v1 · Yuji Takubo, Daniele Gammelli, Marco Pavone et al.2026 · 0 citationsarXiv
Filter-Aware Fine-Tuning for Safe Humanoid Whole-Body Tracking2610.02341v1 · Pranit Mohnot, Christian Helten, Daniele Gammelli et al.2026 · 0 citationsarXiv
DeepJEPA: Scaling World Models from Within2610.00368v1 · Zijian Jin, Yunbei Zhang, Yuanzhe Liu et al.2026 · 0 citationsarXiv
LIBERO-MAX: Do Robot Policies Adapt When the World Changes?2609.36518v1 · Yunbei Zhang, Zijian Jin, Yuanzhe Liu et al.2026 · 0 citationsarXiv
Efficiently Approximating Attention Is Hard2609.37261v1 · Lukas Haverbeck, Carmen Amo Alonso, Andres Felipe Posada-Moreno et al.2026 · 0 citationsarXiv
Vision-Language-Action Autonomous Driving Agent with Language-based Memory2609.38641v1 · Kai Yan, Xiangyu Chen, Yulong Cao et al.2026 · 0 citationsarXiv
SCOPD: Sparse-Context On-Policy Self-Distillation for Efficient Vision-Language Models2609.34044v1 · Ahmadreza Jeddi, Enming Zhang, Jasper Gerigk et al.2026 · 0 citationsarXiv
Structured World-State Reasoning for Agentic Robotic Search2609.23841v1 · Finley R. Holt, Luis A. Pabon, John Irvin Alora et al.2026 · 0 citationsarXiv
A Scene Language Model for Open-Vocabulary Scene Mapping2609.21400v1 · Adam Lilja, Fabio Hubel, Siming He et al.2026 · 0 citationsarXiv
OPTED: On-Policy Fine-Tuning for End-to-End Driving using a Render-Free Teacher2609.20756v1 · Damiano Da Col, Maximilian Igl, Peter Karkus et al.2026 · 0 citationsarXiv
ElastiQP: An Always-Feasible QP Solver for Constrained Robot Control2609.19080v1 · Daniel Morton, Jon Arrizabalaga, Zachary Manchester et al.2026 · 0 citationsarXiv
Coverage Aware Active Evaluation for Failure Discovery with Paired Systems2608.13719v1 · Anjali Parashar, Rachel Luo, Apoorva Sharma et al.2026 · 0 citationsarXiv
Navigating the Proximity-Safety Balance: Constraint Decomposition for Human Following in Pedestrian Crowds2608.10056v1 · Shiting Gong, Jianpeng Yao, Jinfeng Wang et al.2026 · 0 citationsarXiv
Principles of Robot Autonomy2608.03496v2 · Daniele Gammelli, Joseph Lorenzetti, Katie Luo et al.2026 · 0 citationsarXiv
Towards Spatial Supersensing in the Wild2607.13681v1 · Tianjun Gu, Tianyu Xin, Kuan Zhang et al.2026 · 0 citationsarXiv
OpenLongTail: Generative Scaling of Long-Tail Driving Data2607.09655v1 · Lulin Liu, Nuo Chen, Yan Wang et al.2026 · 0 citationsarXiv
WildCity: A Real-World City-Scale Testbed for Rendering, Simulation, and Spatial Intelligence2607.06838v1 · Xiangyu Han, Mengyu Yang, Jiaqi Li et al.2026 · 0 citationsarXiv
Do Vision-Language-Action Models Mean What They Say? On the Role of Faithfulness in Embodied Reasoning2607.04681v1 · Matthew Foutter, Matteo Cercola, Lena Wild et al.2026 · 0 citationsarXiv
LLM-as-a-Verifier: A General-Purpose Verification Framework2607.05391v2 · Jacky Kwok, Shulu Li, Pranav Atreya et al.2026 · 0 citationsarXiv
The risk of KV cache compression2607.01520v1 · Lukas Haverbeck, Carmen Amo Alonso, Andres Felipe Posada-Moreno et al.2026 · 0 citationsarXiv
A Physics-Grounded Benchmark for Multi-Agent Dynamics in World Models2606.28757v2 · Nuo Chen, Lulin Liu, Zihao Li et al.2026 · 0 citationsarXiv
Pose-Agnostic Robotic Functional Grasping via Observation-Action Canonicalization2606.21148v1 · Le Qiu, Cole Harrison, Jiankai Sun et al.2026 · 0 citationsarXiv
FARM: Find Anything using Relational Spatial Memory2606.15476v3 · Siming He, Leo Huang, Adam Lilja et al.2026 · 0 citationsarXiv
DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?2606.12402v2 · Jadelynn Dao, Milan Ganai, Yasmina Abukhadra et al.2026 · 0 citationsarXiv
Planning-aligned Token Compression for Long-Context Autonomous Driving2606.07464v3 · Zhixuan Liang, Yuxiao Chen, Yurong You et al.2026 · 0 citationsarXiv
X4Val: Learning Neural Surrogates for Variance-Reduced Policy Evaluation2606.05159v1 · Rachel Luo, Michael Watson, Apoorva Sharma et al.2026 · 0 citationsarXiv
Semantic Constraint Synthesis for Adaptive Trajectory Optimization via Large Language Models2606.04123v1 · Eleanor Brosius, Yuji Takubo, Daniele Gammelli et al.2026 · 0 citationsarXiv
Cosmos 3: Omnimodal World Models for Physical AI2606.02800v4 · NVIDIA: Aditi, Niket Agarwal, Arslan Ali et al.2026 · 1 citationarXiv
StressDream: Steering Video World Models for Robust Policy Evaluation and Improvement2606.00267v2 · Junwon Seo, Sushant Veer, Ran Tian et al.2026 · 0 citationsarXiv
Constrained Whole-Body Tracking for Humanoid Robots2606.00374v2 · Daniel Morton, Pranit Mohnot, Marco Pavone2026 · 0 citationsarXiv
Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving2605.23163v2 · Kewei Zhang, Jin Wang, Sensen Gao et al.2026 · 0 citationsarXiv
VISTAQA: Benchmarking Joint Visual Question Answering and Pixel-Level Evidence2605.20676v1 · Mozhgan Nasr Azadani, Yimu Wang, Yongpeng Zhu et al.2026 · 0 citationsarXiv
Bridging Structure and Language: Graph-Based Visual Reasoning for Autonomous Road Understanding2605.20942v1 · Lena Wild, Katie Z Luo, Marco Pavone2026 · 0 citationsarXiv
Vision-Based Safe Human-Robot Collaboration with Uncertainty Guarantees2604.15221v3 · Jakob Thumm, Marian Frei, Tianle Ni et al.2026 · 0 citationsarXiv
frax: Fast Robot Kinematics and Dynamics in JAX2604.04310v2 · Daniel Morton, Marco Pavone2026 · 0 citationsarXiv
Bilevel MPC for Linear Systems: A Tractable Reduction and Continuous Connection to Hierarchical MPC2603.29265v2 · Ryuta Moriyasu, Carmen Amo Alonso, Marco Pavone2026 · 0 citationsarXiv
Learning Actuator-Aware Spectral Submanifolds for Precise Control of Continuum Robots2603.23044v1 · Paul Leonard Wolff, Hugo Buurmeijer, Luis Pabon et al.2026 · 1 citationarXiv
: An Open Foundation Model Towards Universal Humanoid Loco-Manipulation2603.12263v1 · Songlin Wei, Hongyi Jing, Boqian Li et al.2026 · 3 citationsarXiv
Observing and Controlling Features in Vision-Language-Action Models2603.05487v2 · Hugo Buurmeijer, Carmen Amo Alonso, Aiden Swann et al.2026 · 0 citationsarXiv
Graph Neural Model Predictive Control for High-Dimensional Systems2602.17601v2 · Patrick Benito Eberhard, Luis Pabon, Daniele Gammelli et al.2026 · 0 citationsarXiv
Career total: at least 655 works. 397 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.