Sakana AI Introduces SAIL for Test-Time Robot Trajectory Scaling
Test-time scaling, proven in language reasoning, may transfer to robot control without touching model weights — SAIL is an early sign of that shift. Robotics researchers and embodied-AI builders can apply the simulator-plus-VLM-feedback loop to their own stacks, trading compute for reliability. The work competes conceptually with end-to-end approaches like GPT-6 Astra's direct robot operation rather than replacing model training.
Sakana AI and the University of Tokyo will present SAIL (Scaling In-Context Imitation Learning) at IROS 2026.
The method uses a policy VLM to generate robot trajectories from a few demonstrations, tests them in simulation, and has an evaluation VLM review videos to flag stalled progress.
Monte Carlo tree search explores and refines alternative trajectories before only the selected one runs on the physical robot.
Across six simulated manipulation tasks, raising the search budget from 1 to 45 candidates improved the average success rate from 25% to 73%.
The team also validated SAIL on a physical robot.
Introducing "Scaling In-Context Imitation Learning" (SAIL) to be presented at #IROS2026. This work is a collaboration between Sakana AI and the University of Tokyo.
Blog: https://t.co/b6xgwIMkUP
What does a robot need before it can tackle a new task?
Teaching a robot something new usually starts with collecting demonstrations and training a policy. But foundation models have already learned from vast amounts of images, text, and robotics-related data. We wanted to see how much of that knowledge we could draw out for robot control without changing the model itself.
Recent demonstrations suggest that GPT-6 Astra can operate physical robots alongside its general language and vision capabilities. Earlier work has also shown that LLMs/VLMs can generate entire sequences of robot movements from a few demonstrations.
However, a foundation model does not necessarily produce a reliable robot trajectory in a single generation. Performance depends on the context provided, and a small error in a movement target can cause the entire task to fail.
We propose SAIL, a method for more reliable VLM-based robot trajectory generation through test-time scaling.
SAIL uses a policy VLM as a robot trajectory generator, conditioned on a few successful demonstrations. It tests the generated trajectory in a simulator and uses an evaluation VLM to review the resulting video and identify where progress stalled. The policy VLM then uses this feedback to revise the trajectory, with Monte Carlo tree search (MCTS) exploring alternatives while refining promising candidates. Only the selected trajectory is sent to the physical robot.
Across six manipulation tasks in simulation, increasing the search budget from one candidate to 45 raised the average rate of finding a successful trajectory from 25% to 73%. We also evaluated SAIL on a physical robot. Our results suggest that robot trajectory generation can benefit from test-time scaling, with additional computation enabling the model to test and refine its proposed actions in simulation.
We think there is more to learn about what existing models can do with this kind of feedback, and how far those improvements carry over to physical robots.
Paper: https://t.co/robokip5Nc 🐟
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Sakana AI, SAIL, robotics, test-time scaling.
Source attribution: curated from Sakana AI (@SakanaAILabs).
