Time is Not a Label: Continuous Phase Rotation for Temporal Knowledge Graphs and Agentic Memory

2026-04-13

AI Summary

This work addresses a key limitation in existing structured memory systems, which model time as discrete labels and consequently struggle to distinguish between persistent and dynamic facts, often erroneously overwriting outdated knowledge. To overcome this, the authors propose RoMem, a novel module that introduces continuous phase rotation in complex vector space to geometrically encode temporal knowledge graphs, allowing obsolete facts to naturally decay without explicit deletion. Additionally, a semantic velocity gating mechanism is designed to predict the temporal volatility of relations from their textual embeddings, dynamically modulating their evolution rates. Experiments demonstrate that RoMem achieves an MRR of 72.6% on ICEWS05–15, improves MRR and accuracy by 2–3× on MultiTQ tasks, leads on the LoCoMo benchmark, preserves static memory intact, and enables zero-shot generalization on FinTMMBench.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Planning, Routing, and Scheduling: Temporal Planning

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Structured memory representations such as knowledge graphs are central to autonomous agents and other long-lived systems. However, most existing approaches model time as discrete metadata, either sorting by recency (burying old-yet-permanent knowledge), simply overwriting outdated facts, or requiring an expensive LLM call at every ingestion step, leaving them unable to distinguish persistent facts from evolving ones. To address this, we introduce RoMem, a drop-in temporal knowledge graph module for structured memory systems, applicable to agentic memory and beyond. A pretrained Semantic Speed Gate maps each relation's text embedding to a volatility score, learning from data that evolving relations (e.g., "president of") should rotate fast while persistent ones (e.g., "born in") should remain stable. Combined with continuous phase rotation, this enables geometric shadowing: obsolete facts are rotated out of phase in complex vector space, so temporally correct facts naturally outrank contradictions without deletion. On temporal knowledge graph completion, RoMem achieves state-of-the-art results on ICEWS05-15 (72.6 MRR). Applied to agentic memory, it delivers 2-3x MRR and answer accuracy on temporal reasoning (MultiTQ), dominates hybrid benchmark (LoCoMo), preserves static memory with zero degradation (DMR-MSC), and generalises zero-shot to unseen financial domains (FinTMMBench).

Problem

• temporal knowledge graphs
• structured memory
• time modeling
• fact evolution
• agentic memory

Innovation

• continuous phase rotation
• temporal knowledge graph
• geometric shadowing
• semantic speed gate
• agentic memory