Agent / long-term memory

Engrama

EngramaBench: Evaluating Long-Term Conversational Memory with Structured Graph Retrieval

On the frontierfirst seen Apr 23, 2026

Current frontier — recent, not yet superseded in the knowledge base

0 papers critique it · 0 beat it on benchmarks

What to use instead

Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.