Neurosymbolic Creative AI

Code for the literature search underpinning the paper "Creativity in Human-AI Systems: A Neurosymbolic Architecture for Constraint Modulation".

The paper argues that creative capacity in human-AI systems comes from an engineered tension between a divergent generation engine and a neurosymbolic convergent constraint layer, drawing on psychedelic neuroscience (precision relaxation, REBUS) as a mechanistic analogy, and grounds this in a PRISMA-ScR-guided scoping review across neurosymbolic AI, cognitive architecture, uncertainty/exploration, ensemble methods, and psychedelic neuroscience.

Licence

MIT License. See LICENSE for details.

Citation

To cite this work, cite the paper:

@misc{cajas2026creativity,
  author       = {Cajas, Sebastian and Dobek, Kacper and Krawiec, Krzysztof and
                   Lunde, Torleif Markussen and Abulibdeh, Rawan and
                   Nageswaran, Neha and Giancotti, Raffaele and Arslan, Janan
                   and Mwavu, Rogers and Patel, Milit and Kleinlein, Ricardo
                   and Celi, Leo Anthony},
  title        = {Creativity in Human-{AI} Systems: A Neurosymbolic
                   Architecture for Constraint Modulation},
  year         = {2026},
  month        = jun,
  howpublished = {SSRN},
  url          = {https://ssrn.com/abstract=7022759},
  doi          = {10.2139/ssrn.7022759}
}