Memory isn't enough. AI needs experience.

Описание и применение

Система <omnimem> добавляет к памяти агента модель опыта, учитывая результаты, усилия и стоимость, что позволяет избежать повторных ошибок и снизить затраты на выполнение задач.

Задача

AI agents rely on memory that stores events but lacks context about cost, effort, and outcomes, leading to repeated dead ends and inefficient learning.

Результат

In ten paired runs, hooks prevented all attempts at abandoned approaches and avoided failing test suites; runs with hooks cost 27% less. On LongMemEval, version 5.x scored 65.6% across 500 questions.

Рабочий процесс

• 1. Record each interaction with outcome, effort, and cost.
• 2. Apply experience scoring to weight memories by effort (1.0×–1.8×).
• 3. Store failures in a graveyard with warnings for future runs.
• 4. Manage memory lifecycle: active → deprioritised → archived → deleted, with recency decay.
• 5. Detect contradictions and deduplicate writes.
• 6. Compile domain skills into SKILL.md from reinforced lessons.
• 7. Use hooks to inject briefings, block abandoned approaches, and record session ends.

Доказательства

With hooks, every attempt at the abandoned approach was stopped before it ran, and no run reached a failing test suite.

Runs with hooks cost 27% less.

On recall, I ran <omnimem> against LongMemEval earlier this year: 5.x scored 65.6% across 500 questions with a GPT-4o judge.

Источник