Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents

TL;DR

This paper introduces a dual process memory architecture for large language models that effectively manages long-horizon scientific workflows by decoupling immediate and long-term memory, enabling sustained performance beyond context limits.

Contribution

The paper presents a novel dual process memory system tailored for scientific agents, addressing long-term knowledge retention and reasoning across extended workflows, outperforming traditional models.

Findings

Maintains 70-85% accuracy with 62% fewer tokens at 10,000 messages.

Architecture excels at numeric/temporal queries and historical retrieval, showing complementary strengths.

Scales to profiles with over 14,000 facts, demonstrating long-term knowledge management.

Abstract

As Large Language Models (LLMs) evolve into persistent scientific collaborators, context window saturation has emerged as a critical bottleneck. Scientific workflows involving iterative data analysis and hypothesis refinement rapidly saturate even extended contexts with dense technical content, while monolithic approaches suffer from quadratic cost scaling and cognitive degradation. We evaluate a Dual Process Memory Architecture that decouples immediate episodic needs (constant 10-message window) from long-term consolidated knowledge (growing at approximately 3 tokens/message). Unlike prior social agent memory systems, our domain-specific consolidation addresses contradictory parameter evolution, multi-hop reasoning across experimental phases, and precise technical fact retention. Through large-scale evaluation spanning 15,000 messages with cross-model validation across six LLMs from three families (OpenAI, Anthropic, Google), totaling 1,440 queries, we establish three key findings. First, while full-context models fail at 10,000 messages due to context overflow, our system maintains 70-85% accuracy with 1-2 second latency using 62% fewer tokens (45,434 vs 120,000+ limit). Second, cross-model validation reveals architecture-level trade-offs independent of specific LLMs: Dual Process excels at numeric/temporal queries (65-90% accuracy) while RAG excels at historical retrieval (60-85%), suggesting complementary deployment strategies. Third, we identify a "Sim-to-Real" gap where synthetic tests maintain constant memory but realistic workflows exhibit linear growth (about 3 tokens/message), with consolidation quality emerging as the primary scalability bottleneck. The architecture successfully manages profiles with 14,000+ scientific facts (125k tokens), demonstrating that domain-specific memory consolidation enables sustained operation beyond full-context limits.