Dynin-Robotics foundation model for scene prediction and robot control
Overview
A research team from Seoul National University College of Engineering has developed Dynin-Robotics, a robotic foundation model that can understand human instructions, predict future scenes, and generate actions to complete tasks. The model outperforms existing models in real-world robot manipulation tasks, achieving an average success rate of 78.4%. The paper has been accepted to NeurIPS 2026 and demonstrates the potential of integrating instruction understanding and scene prediction for robotic control.
From Tech Xplore — Robotics
Event evidence and corrections
1 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · Reported success rate: 78.4 percent · Basis not reported · Differing source assertions
“the target scene should look like once a task is completed, and use these predictions to generate and select actions. In evaluations across four real-world robot manipulation settings, Dynin-Robotics achieved an average success rate of 78.4%, outperforming the comparison models from NVIDIA and U.S.-based robot AI company Physical Intelligence used in the experiments. The paper describing the research has been accepte”
Exact source · revision 2science-x
Reported quantity · Reported success rate: 73 percent · Basis not reported · Differing source assertions
“e team evaluated four manipulation settings: moving fruit, sorting blocks by color, stacking blocks without a required order and stacking blocks in a specified color sequence. Averaged across the four settings, NVIDIA's "GR00T N1.6" achieved a success rate of 73.0%, NVIDIA's "Cosmos Policy" achieved 73.6%, and Physical Intelligence's "π0.5" achieved 76.9%. Dynin-Robotics recorded 78.4%, the highest average success ra”
Exact source · revision 2science-x
Reported quantity · action-generation speed improvement: 29.2 other · Basis not reported · Differing source assertions
“sequence of actions and predicts it in parallel. Information repeatedly used during computation is also reused to improve the efficiency of generating the robot's next action. The model's internal action-generation speed improved by up to approximately 29.2 times compared with the unoptimized version and was up to approximately 36.6 times faster than Physical Intelligence's "π0.5." In real-world robot experiments, th”
Exact source · revision 2science-x
Reported quantity · action-generation speed improvement: 36.6 other · Basis not reported · Differing source assertions
“sequence of actions and predicts it in parallel. Information repeatedly used during computation is also reused to improve the efficiency of generating the robot's next action. The model's internal action-generation speed improved by up to approximately 29.2 times compared with the unoptimized version and was up to approximately 36.6 times faster than Physical Intelligence's "π0.5." In real-world robot experiments, th”
Exact source · revision 2science-x
Evidence dependency changes (2)
- 2026-10-30T22:24:57.234Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T22:24:38.337Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
Report timeline
Follow attributed reports and material updates.
- Tech Xplore — RoboticsSignalRobot AI predicts future scenes to guide its next moves
A research team from Seoul National University College of Engineering has developed Dynin-Robotics, a robotic foundation model that can understand human instructions, predict future scenes, and generate actions to complete tasks. The model outperforms existing models in real-world robot manipulation tasks, achieving an average success rate of 78.4%. The paper has been accepted to NeurIPS 2026 and demonstrates the potential of integrating instruction understanding and scene prediction for robotic control.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).