Prism: An Evolutionary Memory Substrate for Multi-Agent Open-Ended Discovery

TL;DR

Prism is a unified evolutionary memory system for multi-agent AI that enhances open-ended discovery through novel memory stratification, causal graph, and evolutionary dynamics, demonstrating significant improvements on benchmarks.

Contribution

It introduces five novel mechanisms—entropy-gated stratification, causal memory graph, value-of-information retrieval, heartbeat-driven consolidation, and replicator-decay dynamics—within a unified decision-theoretic framework.

Findings

Achieves 88.1 LLM-as-a-Judge score on LOCOMO benchmark, outperforming previous methods.

Four-agent Prism outperforms single-agent baselines by 2.8 times in evolutionary optimization tasks.

Demonstrates convergence to an Evolutionary Stable Memory Set (ESMS).

Abstract

We introduce \prism{} (\textbf{P}robabilistic \textbf{R}etrieval with \textbf{I}nformation-\textbf{S}tratified \textbf{M}emory), an evolutionary memory substrate for multi-agent AI systems engaged in open-ended discovery. \prism{} unifies four independently developed paradigms -- layered file-based persistence, vector-augmented semantic memory, graph-structured relational memory, and multi-agent evolutionary search -- under a single decision-theoretic framework with eight interconnected subsystems. We make five contributions: (1)~an \emph{entropy-gated stratification} mechanism that assigns memories to a tri-partite hub (skills/notes/attempts) based on Shannon information content, with formal context-window utilization bounds; (2)~a \emph{causal memory graph} $\mathcal{G} = (V, E_r, E_c)$ with interventional edges and agent-attributed provenance; (3)~a \emph{Value-of-Information retrieval} policy with self-evolving strategy selection; (4)~a \emph{heartbeat-driven consolidation} controller with stagnation detection via optimal stopping theory; and (5)~a \emph{replicator-decay dynamics} framework that interprets memory confidence as evolutionary fitness, proving convergence to an Evolutionary Stable Memory Set (ESMS). On the LOCOMO benchmark, \prism{} achieves 88.1 LLM-as-a-Judge score (31.2\% over Mem0). On CORAL-style evolutionary optimization tasks, 4-agent \prism{} achieves 2.8$\times$ higher improvement rate than single-agent baselines.%