# WorldDB: A Vector Graph-of-Worlds Memory Engine with Ontology-Aware Write-Time Reconciliation

Source: [WorldDB: A Vector Graph-of-Worlds Memory Engine with Ontology-Aware Write-Time Reconciliation](<https://www.sota2.com/research/paper/69e7052d7d0172140ec7f35b>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-22T05:23:53.567Z

Authors: Harish Santhanalakshmi Ganesan

Published: 2026-04-20T16:30:53.000Z

## Abstract

Persistent memory is the bottleneck separating stateless chatbots from long-running agentic systems. Retrieval-augmented generation (RAG) over flat vector stores fragments facts into chunks, loses cross-session identity, and has no first-class notion of supersession or contradiction. Recent bitemporal knowledge-graph systems (Graphiti, Memento, Hydra DB) add typed edges and valid-time metadata, but the graph itself remains flat: no recursive composition, no content-addressed invariants on nodes, and edge types carry no behavior beyond a label. We present WorldDB, a memory engine built on three commitments: (i) every node is a world -- a container with its own interior subgraph, ontology scope, and composed embedding, recursive to arbitrary depth; (ii) nodes are content-addressed and immutable, so any edit produces a new hash at the node and every ancestor, giving a Merkle-style audit trail for free; (iii) edges are write-time programs -- each edge type ships on\_insert/on\_delete/on\_query\_rewrite handlers (supersession closes validity, contradicts preserves both sides, same\_as stages a merge proposal), so no raw append path exists. On LongMemEval-s (500 questions, ~115k-token conversational stacks), WorldDB with Claude Opus 4.7 as answerer achieves 96.40% overall / 97.11% task-averaged accuracy, a +5.61pp improvement over the previously reported Hydra DB state-of-the-art (90.79%) and +11.20pp over Supermemory (85.20%), with perfect single-session-assistant recall and robust performance on temporal reasoning (96.24%), knowledge update (98.72%), and preference synthesis (96.67%). Ablations show that the engine's graph layer -- resolver-unified entities and typed refers\_to edges -- contributes +7.0pp task-averaged independently of the underlying answerer.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Long-context conversation memory reasoning on LongMemEval-s oracle variant HuggingFace xiaowu0162](<https://www.sota2.com/research/sota/long-context-conversation-memory-reasoning-on-longmemeval-s-oracle-variant-huggingface-xiaowu0162.md>): Long-context conversation memory reasoning; LongMemEval-s oracle variant HuggingFace xiaowu0162; Single-session Accuracy (User): 98.57.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.18478>)

[PDF](<https://arxiv.org/pdf/2604.18478.pdf>)