ColaBench Benchmark for Memory-Guided Knowledge Acquisition in Personal Agents
October 23, 2026
ColaBench evaluates how personal agents retrieve and apply external knowledge based on long-term user history. The synthetic benchmark features 349 tasks across 10 profiles and a 13,033-document knowledge base, using an Evidence Graph to connect personal conditions to specific document disclosures.
HOW THIS AFFECTS YOU
●
builderThis provides a framework for testing the personalization capabilities of your user-facing agents.
●
researcherYou can use this to evaluate how well models integrate historical context with new knowledge.