TreeNets (Decision Tree Based Neural Networks)

Independent researcher Younes Naghibi released a paper introducing TreeNets, a method that reduces large language model size by combining neural networks with decision trees. The approach includes a TreeNet-based BART model on Hugging Face, aiming to improve efficiency and interpretability.

My current research paper introduces so called TreeNets, which can drastically reduce the size of LLMs. In general, TreeNets help us understand the fundamental connection between neural networks and decision trees. You can find my paper here https://codeberg.org/ynaghibi/TreeNets/src/branch/main/TreeNets.pdf . Or go to my repository https://codeberg.org/ynaghibi/TreeNets/src/branch/main directly. This is the first draft of my paper, so it will require changes in the future. I have also uploaded a LLM based on TreeNets and Bart. You can find it here https://huggingface.co/ynaghibi/tree bart xsum 6 6 . Please let me know if there are any problems with the uploads. I am an independent researcher and this is my first publication.