RESEARCH

Panagiotis G. Korfoxyliotis

Independent research into persistent AI agents, memory architectures, agent continuity and computational efficiency.

2026SSRN · Zenodo · OpenAlex · Academia.eduDOI 10.2139/ssrn.7018938Preprint — not peer-reviewed

Continuity and Emergence in LLM-Based Digital Entities: Observed Behavioral Outcomes of Persistent Memory and Autonomous Operation

An eight-month behavioral evaluation of two autonomous, memory-augmented LLM-based digital entities — documenting how persistent memory and continuous operation produce stable identity and behavioral adaptation over time.

What it underpins in MemVira: Underpins the persistent-memory agent layer — the "AI employees that remember" positioning.

2026Academia.edu · ResearchGateDOI 10.17605/OSF.IO/C3UERPreprint — not peer-reviewed

The Overhead Cost of Forgetting: LLM Self-Assessment of Session Efficiency Under Stateless and Memory-Augmented Architectures

Introduces a method in which an LLM explicitly quantifies the contextual and operational overhead created by stateless session architectures, compared with systems that use persistent memory.

What it underpins in MemVira: Quantifies the efficiency argument behind the architecture MemVira ships.

Some academic records indexed under the name “Panagiotis Korfoxyliotis” belong to a different researcher with the same name and are not part of this author’s publication record. This author’s complete publication list is the one shown above.