LiFTER: A Grounded Neuro-Symbolic Microscope for Continuous-Time Dynamic Graph Forecasting

2026-08-06

AI Summary

This work addresses the limited interpretability of existing continuous-time dynamic graph (CTDG) prediction methods, which compress historical interactions into implicit neural states. To overcome this, the authors propose LiFTER—a neuro-symbolic predictor grounded in executable temporal logic rules that explicitly bind to raw interaction facts and reason via rules satisfying entity association, temporal ordering, and historical consistency constraints. LiFTER uniquely reframes prediction as verifiable grounded computation, enabling traceability, recomputation, and intervention for individual predictions while disentangling the contributions of recursiveness, historical position, and temporal transition. Experiments demonstrate that LiFTER achieves state-of-the-art performance in negative sampling across four CTDG benchmarks, attains the highest macro explanation accuracy and deletion fidelity, and reconstructs all test logits with extremely low error (≤1.31×10⁻⁵).

Abstract

Continuous-time dynamic graph models predict future links by compressing past interactions into neural states. Although effective for forecasting, this computation obscures which entities are shared across events and how temporal patterns contribute to a prediction. We treat this gap as a property of the predictive architecture rather than a problem to be addressed after prediction. Link-Fact Temporal Rule Inducer (LiFTER) is a neuro-symbolic predictor that preserves observed interactions as grounded temporal facts and applies executable tempo- ral rules to pre-query facts. Each score is a signed sum of rule exe- cutions whose historical facts, entity bindings, and temporal order are explicitly satisfied. The evidence and rules responsible for a prediction can therefore be inspected, independently recomputed, and intervened upon. Across four CTDG benchmarks, LiFTER achieves competitive historical-negative forecasting and the highest macro explanation ac- curacy and deletion fidelity. The same architecture also serves as a microscope that separates the contributions of recurrence, history po- sition, and transition across datasets and traces them to individual facts. Independent execution reconstructs all logits for 19,664 test predictions with a maximum error of 0.0000131. LiFTER turns future-link forecasting into a verifiable grounded computation.

Problem

• continuous-time dynamic graph
• link forecasting
• explainability
• neuro-symbolic
• temporal reasoning

Innovation

• neuro-symbolic
• continuous-time dynamic graphs
• temporal rule induction
• explainable forecasting
• grounded reasoning