- Verses AI’s new robotics model performs complex household tasks like tidying, grocery prep, and table setting without any pre-training, unlike deep learning models that require billions of training steps
- The model uses active inference to dynamically adapt to changing environments, outperforming traditional robotics systems that often freeze when faced with unexpected scenarios
- In comparative tests, Verses’ model achieved a 66.5 per cent success rate, which is higher than the best deep learning alternative
- Verses AI stock (CBOE:VERS) opened trading at C$8.58
Cognitive computing company Verses AI Inc. (CBOE:VERS) has announced a major advancement in robotics with the unveiling of its new agentic robotics architecture, capable of performing complex household tasks without any pre-training. The model, which leverages a hierarchical active inference system, has demonstrated superior performance compared to traditional deep learning-based robotics systems.
This content has been prepared as part of a partnership with Verses AI Inc. and is intended for informational purposes only.
In a peer-reviewed paper titled “mobile manipulation with active inference for long-horizon rearrangement tasks”, researchers from Verses detail how their model successfully completed three long-horizon tasks—tidying a room, preparing groceries, and setting a table—with a success rate of 66.5 per cent, outperforming the previous best deep learning alternative which achieved 54.7 per cent but required 1.3 billion training steps.
Unlike conventional robotics systems that rely on either rigid pre-programming (drive-by-wire) or massive datasets for deep learning, Verses’ model adapts dynamically to its environment. This adaptability is crucial in real-world settings where unexpected changes—like a misplaced box or a fallen bottle—can halt traditional robots. Verses’ architecture bypasses these limitations by using active inference, a cognitive framework inspired by human perception and decision-making.
The system integrates three core modules:
- Vision: for mapping the environment,
- Planning: for decomposing tasks into manageable steps,
- Control: for executing precise movements.
This design allows the robot to explore and adapt in real time, eliminating the need for extensive pre-programming or retraining when faced with new scenarios.
Sean Wallingford, former CEO of Swisslog, praised the innovation in a news release: “For instance, if you bring a robot to a new factory or ask it to do a different job, it will need a lot of re-training and may not be reliable. Verses breakthroughs are exciting, because they offer an alternative approach. If we can deploy robots without training, they will be viable in a wide range of activities, from factories and warehouses to domestic and commercial applications.”
The implications of this breakthrough are far-reaching. By removing the dependency on costly and time-consuming training, Verses AI’s model opens the door to more flexible, scalable, and reliable robotics solutions. As CTO Hari Thiruvengada added, “Our breakthrough has the potential to transform how robots operate across industries, from factories and warehouses to homes and public spaces, potentially unlocking a new era of truly adaptive, reliable automation.”
Specializing in biologically inspired distributed intelligence, Verses AI is focused on developing next-generation artificial intelligence solutions, including the Genius platform.
Verses AI stock (CBOE:VERS) opened trading more than 4 per cent higher at C$8.58.
Join the discussion: Find out what the Bullboards are saying about Verses AI and check out Stockhouse’s stock forums and message boards.
Stockhouse does not provide investment advice or recommendations. All investment decisions should be made based on your own research and consultation with a registered investment professional. The issuer is solely responsible for the accuracy of the information contained herein. For full disclaimer information, please click here.