Start here: 1) Open `v1. 6ᵣeport. csv` (baseline vs Mnemosyne table) 2) Open `v1. 6ₜerminalₜrace. txt` and search for `REJECT` (fail-closed rollback proof) 3) Inspect `v1. 6ₜraceₚerframe. json` (per-frame, per-attempt trace) Source code (GitHub release): https: //github. com/Mnemosyne-Protocol/Mnemosyne-Core/releases/tag/v1. 6. 0-beta (Abstract + short pitch) Generative AI models (LLMs and diffusion/video models) achieve state-of-the-art results in isolated tasks but suffer from Contextual Fragmentation in temporal workflows, leading to continuity drift and expensive human rework. This paper introduces the Mnemosyne Protocol, a vector-based orchestration layer designed to maintain semantic and visual consistency across heterogeneous generative systems while enforcing Local-First Sovereignty for studio IP. Mnemosyne operationalizes continuity as a discrete-time, fail-closed verification gate (a conjunctive product-of-constraints) evaluated across frame sequences, with an explicit rollback + localized re-sampling mechanism before final rendering. Preliminary simulations demonstrate substantial reductions in continuity hallucination rates under defined constraints. What’s new in (v1. 6. 1 Packaging/Preview Patch) v1. 6. 1 adds reproducible benchmarks/baselines + standardized metrics and publishes executable traces to make the claims falsifiable. Code is licensed under MIT (see LICENSE in the archive) ; the paper text is CC BY 4. 0.
Mert Kerem Salman (Mon,) studied this question.