The new system called 3DGraphlm takes into account not only themselves, but also the spatial and semantic ties. This is particularly important in rooms with a large number of items such as kitchen, workshop or office. The model has been examined in well -known data clusters with the correct 3D reconstruction of real facilities and text descriptions of objects. Vicuna-V1.5 and Llama3 operating on energy-saving devices suitable for robotics are used.
Based on the test results, 3DGraphLLM has exceeded many existing methods, including other approaches related to language models. The model has improved more than 7% of accuracy compared to the basic solution, and worked faster and more economically in sources than the most advanced analogues.
Now a scientist team is engaged in integration of the method into real robot systems. The main task is that robots not only see the objects, but understand their relationships and perform their tasks effectively. These skills are important to create new generation service and home robots.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
