PlugMem: Rethinking Memory for AI Agents

Today's AI agents store long interaction histories but often struggle to reuse them effectively. The accumulation of raw data can overwhelm agents with excessive and low-value context.

PlugMem is a memory module that transforms interaction history into structured, reusable knowledge. This approach overcomes the limitations of traditional memory systems, which often rely on storing large amounts of unstructured text.

How PlugMem Works

PlugMem converts the agent's interaction history (dialogues, documents, web sessions) into structured and compact knowledge units that can be reused in different contexts. The system is based on three main components:

  • Structure: Raw interactions are standardized and transformed into propositional knowledge (facts) and prescriptive knowledge (reusable skills). These knowledge units are organized into a structured memory graph.
  • Retrieval: Instead of retrieving long passages of text, PlugMem retrieves knowledge units aligned with the current task. High-level concepts and inferred intents serve as routing signals, surfacing the most relevant information.
  • Reasoning: The retrieved knowledge is distilled into concise, task-ready guidance before being passed to the base agent, ensuring that only decision-relevant knowledge enters the agent's context window.

Evaluating PlugMem

PlugMem was evaluated on three benchmarks that require different types of memory:

  • Answering questions in long multi-turn conversations.
  • Finding facts spanning multiple Wikipedia articles.
  • Making decisions while browsing the web.

Across all three, PlugMem consistently outperformed both generic retrieval methods and task-specific memory designs, while allowing the AI agent to use a significantly smaller memory token budget.

Toward Reusable Memory for Agents

The goal is for agents to carry useful facts and strategies from one task to the next, rather than starting from scratch each time. PlugMem represents a step in this direction, grounding memory design in cognitive principles and treating knowledge as the primary unit of reuse.