works on

From the 1 of 13 linked papers with an AI index.

collaborators

13 papers

cs.RO2026

AgenticFocus: Object-Preserving Mixed Reality Synthesis from Human FPV Video for Dexterous Humanoid Learning

Iaroslav Kolomiets, Miguel Altamirano Cabrera, Artem Lykov +6

The paper presents AgenticFocus, a mixed-reality pipeline that turns ordinary first-person human videos into robot-ready demonstrations by reconstructing hidden object geometry, co…

cs.RO2026

Output-Level Regularization Eliminates the Seed Lottery in Single-GPU VLA Fine-Tuning

Jeffrin Sam, Dzmitry Tsetserukou

Fine-tuning a vision-language-action model (VLA-JEPA) on a single GPU should be simple: load a pretrained checkpoint, run training, deploy. There is a hidden danger. Run the same f…

cs.RO2026

Action Agent: Agentic Video Generation Meets Flow-Constrained Diffusion

Jeffrin Sam, Nguyen Khang, Yara Mahmoud +2

We present Action Agent, a two-stage framework that unifies agentic navigation video generation with flow-constrained diffusion control for multi-embodiment robot navigation. In St…

cs.RO2026

GenerativeMPC: VLM-RAG-guided Whole-Body MPC with Virtual Impedance for Bimanual Mobile Manipulation

Marcelino Julio Fernando, Miguel Altamirano Cabrera, Jeffrin Sam +3

Bimanual mobile manipulation requires a seamless integration between high-level semantic reasoning and safe, compliant physical interaction - a challenge that end-to-end models app…

cs.RO2026

DiffusionAnything: End-to-End In-context Diffusion Learning for Unified Navigation and Pre-Grasp Motion

Iana Zhura, Yara Mahmoud, Jeffrin Sam +4

Efficiently predicting motion plans directly from vision remains a fundamental challenge in robotics, where planning typically requires explicit goal specification and task-specifi…

cs.RO2026

DreamToNav: Generalizable Navigation for Robots via Generative Video Planning

Valerii Serpiva, Jeffrin Sam, Chidera Simon +4

We present DreamToNav, a novel autonomous robot framework that uses generative video models to enable intuitive, human-in-the-loop control. Instead of relying on rigid waypoint nav…