External Hippocampus: Topological Cognitive Maps for Guiding Large Language Model Reasoning

TL;DR

The paper introduces the External Hippocampus framework, which creates topological cognitive maps to improve reasoning in small language models by guiding information flow and addressing reasoning deadlocks without additional training.

Contribution

It presents a novel topological cognitive mapping approach for language model reasoning, enabling efficient intervention and deadlock resolution during inference.

Findings

Achieves 81.20% accuracy on challenging problems

Reduces reasoning time by at least 15 times

Identifies 'Cognitive Vortex' as a reasoning deadlock pattern

Abstract

This paper proposes the External Hippocampus framework, which models language model reasoning from a cognitive dynamics perspective as the flow of information energy in semantic space. Unlike traditional weight-space optimization methods, this framework constructs topological cognitive maps through dimensionality reduction projection, enabling precise navigation and intervention of energy flow at test time while avoiding substantial computational requirements and demonstrating predictable intervention patterns. The method effectively addresses the cognitive deadlock problem in multi-step reasoning for small models. Experiments on models <=7B parameters show: map-guided methods achieve 81.20% accuracy on 500 challenging problems (relative baseline +16.80%), reduce reasoning time by >= 15x, with key findings revealing that reasoning stagnation manifests as "Cognitive Vortex" and low-entropy potential wells, while temperature perturbations effectively restart energy flow. The framework requires no additional training, possesses autonomous growth capability, and provides an efficient and controllable topological-aware solution for small model reasoning.