Reproduce “From Unstructured Recall to Schema-Grounded Memory: Reliable AI Memory via Iterative, Schema-Aware Extraction” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Alex Petrov, Alexander Gusak, Denis Mukha, Dima Korolev
The authors present xmemory, a persistent external-memory system that stores facts in schema-defined records for later queries. Its write path extracts and validates records, retries failed stages, and preserves unknowns rather than guessing; reads query records as facts change. Across four structured domains, xmemory reaches 97.10% micro-averaged fact-level F1, versus 87.24% for Mem0 without graph, in an LLM-judge product comparison.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.