Experience Compression Spectrum: Unifying Memory, Skills, and Rules in LLM Agents

TL;DR

This paper introduces the Experience Compression Spectrum, a unifying framework that positions memory, skills, and rules along a compression axis to improve efficiency in LLM agents, highlighting gaps and open problems.

Contribution

It proposes a unified spectrum for experience compression in LLM agents, revealing fixed compression levels in existing systems and identifying the need for adaptive, cross-level compression.

Findings

Existing systems operate at fixed compression levels.

Transferability increases with compression but reduces specificity.

Knowledge lifecycle management is largely neglected.

Abstract

As LLM agents scale to long-horizon, multi-session deployments, efficiently managing accumulated experience becomes a critical bottleneck. Agent memory systems and agent skill discovery both address this challenge -- extracting reusable knowledge from interaction traces -- yet a citation analysis of 1,136 references across 22 primary papers reveals a cross-community citation rate below 1%. We propose the \emph{Experience Compression Spectrum}, a unifying framework that positions memory, skills, and rules as points along a single axis of increasing compression (5--20$\times$ for episodic memory, 50--500$\times$ for procedural skills, 1,000$\times$+ for declarative rules), directly reducing context consumption, retrieval latency, and compute overhead. Mapping 20+ systems onto this spectrum reveals that every system operates at a fixed, predetermined compression level -- none supports adaptive cross-level compression, a gap we term the \emph{missing diagonal}. We further show that specialization alone is insufficient -- both communities independently solve shared sub-problems without exchanging solutions -- that evaluation methods are tightly coupled to compression levels, that transferability increases with compression at the cost of specificity, and that knowledge lifecycle management remains largely neglected. We articulate open problems and design principles for scalable, full-spectrum agent learning systems.