SEMA is an experimental AI architecture that investigates whether language understanding and reasoning can emerge from explicit semantic structure rather than next-token prediction. It serves as a non-parametric, instance-based reasoning system utilizing Vector Symbolic Architecture (VSA/HDC) over content-addressable memory.
- CPU-first design with multilingual support by construction.
- Uses transparent internal representations instead of opaque embeddings.
- Implements explicit relational memory rather than storing knowledge solely in neural weights.
- Employs composable semantic representations via VSA/HDC.
- Utilizes graph-based reasoning instead of autoregressive generation.
The project is source available to allow researchers to study, reproduce, and challenge the architecture while protecting independent research continuity.