A dark-lucid reinforcement learning architecture that learns, dreams, and acts through latent world models, causal verification, and memory-anchored adaptation under uncertainty.
-
Updated
Mar 23, 2026 - Jupyter Notebook
A dark-lucid reinforcement learning architecture that learns, dreams, and acts through latent world models, causal verification, and memory-anchored adaptation under uncertainty.
Hallucination-free personal AI with near-infinite memory & context. Every output is verified, not believed. FRT 366 causal-verification kernel + TCS long-context memory.
Causal Free Network Service — bring your own VPS, fill in your node, go. Secure tunnel client with a causal security kernel and a network-adaptive firewall.
End-to-end prime factorization in a generative LM. 40M-param GPT that learns algebraically verifiable prime-factor signatures at negligible language cost (+1.7% PPL). Paper (Zenodo) + triadic-head (PyPI) + reptimeline.
A method-neutral protocol for recording and verifying evidence that latent model states causally influenced downstream decisions, outputs, or actions, without requiring storage of raw internal activations.
Does your AI agent actually follow rules? 13 pre-registered experiments + 5-layer verification architecture. Paper, data, code — all public.
To associate your repository with the causal-verification topic, visit your repo's landing page and select "manage topics."