GRAVITY: Architecture-Agnostic Structured Anchoring for Long-Horizon Conversational Memory

TL;DR

GRAVITY introduces a structured memory module that extracts relational, temporal, and thematic representations from conversations, enhancing long-horizon conversational agents without altering their architecture.

Contribution

It provides a plug-and-play method to inject structured knowledge representations into prompts, improving reasoning in conversational agents across diverse memory systems.

Findings

GRAVITY improves LLM-judge accuracy by 7.5-10.1%.

Weak baseline systems benefit the most, with up to 12.2% improvement.

The approach is architecture-agnostic and effective across multiple benchmarks.

Abstract

Long-horizon conversational agents rely on memory systems with increasingly sophisticated retrieval mechanisms. However, retrieved fragments are typically fed to the language model as unstructured text, lacking the relational, temporal, and thematic structures essential for complex reasoning. To bridge this reasoning gap, we introduce GRAVITY (\textbf{G}eneration-time \textbf{R}elational \textbf{A}nchoring \textbf{V}ia \textbf{I}njected \textbf{T}opological Memor\textbf{Y}), a plug-and-play structured memory module. GRAVITY extracts three complementary knowledge representations from raw conversational utterances: entity profiles grounded in relational graphs, temporal event tuples linked into causal traces, and cross-session topic summaries. At generation time, it injects these representations into the host system's prompt as structured anchoring contexts. This approach effectively synthesizes scattered evidence into a coherent, query-relevant context without requiring any architectural modifications to the host model. Extensive evaluations across five diverse memory systems on the LongMemEval and LoCoMo benchmarks demonstrate the efficacy of our approach. On average, GRAVITY improves LLM-judge accuracy by 7.5--10.1%. Gains are inversely correlated with baseline strength: the weakest host improves by 12.2% while the strongest still gains 3.8--5.7%. These findings establish structured context anchoring as a broadly effective, architecture-agnostic augmentation paradigm for long-horizon conversational memory.