Seeking arXiv cs.AI endorsement — neuroscience-inspired memory architecture for AI agents

Hi everyone,

I’m an independent researcher (Zensation AI) seeking endorsement for my first arXiv submission in cs.AI.

Paper: “ZenBrain: A Neuroscience-Inspired 7-Layer Memory Architecture for Autonomous AI Systems”

Summary: ZenBrain is the first AI memory system grounded in cognitive neuroscience. It implements 7 memory layers (working, short-term, episodic, semantic, procedural, core, cross-context) with 12 algorithms including Hebbian learning, FSRS spaced repetition, sleep-time consolidation (Stickgold & Walker 2013), and Bayesian confidence propagation.

Prior art: Published as defensive publication on TDCommons (dpubs_series/9683) and archived on Zenodo (DOI: 10.5281/zenodo.19353663). Open-source npm packages with 9,000+ tests.

Why this matters: Recent surveys (arxiv:2603.07670) identify “deeper neuroscience integration” as a key open challenge in AI agent memory. No existing system (MemGPT, Mem0, A-Mem, Zep) implements neuroscience-grounded mechanisms. ZenBrain addresses this gap.

I’d be very grateful for endorsement in cs.AI. Happy to share the full manuscript.

Best regards,

Alexander Bering

Zensation AI