ResearchMemory as Ontology Constitutional Memory Architecture for Persistent AgentsarXiv·high signalXBlueskyLinkedInCopy linkWhen agent models are swappable but identity must persist, memory IS the ontological foundation. Four-layer governance hierarchy for agent identity.SourceSource pagearXiv↳ Follow the threadStack layer / ContrastContinuity Kernel: a transactional commit protocol for long-lived agent state, model-checked over 2.8M statesarXivStack layer / ContrastTwo-Tier 'Librarian + Writer' Architecture Removes 6,845 Cross-Section Contradictions From LLM Research Reports to ZeroarXiv 2608.12984Policy dependency / Stack layerGive each agent role its own LoRA expert instead of just a different promptarXiv 2608.09251Policy dependency / Stack layerExport Controls Turn Frontier Model Access Into a Cyber-Defence Dependency for Middle PowersarXiv 2608.13272Stack layer / ContrastInterSAGE Proposes the Trust Layer MCP, A2A and ANP Left Out of the Internet of AgentsarXiv 2608.13030Stack layer / Update threadLLMs Systematically Self-Identify as 'Highly Similar' to Other Models — and It Changes Whether They CooperatearXiv 2608.12125Stack layer / Follow-up threadMechanist: An Agentic System for Autonomous Interpretability Research, Built on a 13,000-Paper Knowledge Graph Over 43M PapersarXiv 2608.12036Stack layer / Update threadPlayWorld Puts Multi-Modal 'Agent Players' Inside World Models Across 171 Scenarios — and Finds Long-Horizon State Persistence Still BrokenarXiv (via HuggingFace Daily Papers)