Search papers, labs, and topics across Lattice.
This paper introduces the FinSurvival Challenge 2025, a benchmark for temporal Web3 intelligence using 21.8 million transaction records from the Aave v3 protocol to model user behavior transitions. The challenge operationalized 16 survival prediction tasks, focusing on real-world temporal dynamics like censoring and non-stationarity over extended horizons. Results showed that domain-aware temporal feature construction outperformed generic modeling approaches, demonstrating the importance of domain expertise in Web3 analytics.
Domain-specific feature engineering still trumps generic modeling when predicting user behavior in Web3, even with millions of transaction records.
Temporal Web analytics increasingly relies on large-scale, longitudinal data to understand how users, content, and systems evolve over time. A rapidly growing frontier is the \emph{Temporal Web3}: decentralized platforms whose behavior is recorded as immutable, time-stamped event streams. Despite the richness of this data, the field lacks shared, reproducible benchmarks that capture real-world temporal dynamics, specifically censoring and non-stationarity, across extended horizons. This absence slows methodological progress and limits the transfer of techniques between Web3 and broader Web domains. In this paper, we present the \textit{FinSurvival Challenge 2025} as a case study in benchmarking \emph{temporal Web3 intelligence}. Using 21.8 million transaction records from the Aave v3 protocol, the challenge operationalized 16 survival prediction tasks to model user behavior transitions.We detail the benchmark design and the winning solutions, highlighting how domain-aware temporal feature construction significantly outperformed generic modeling approaches. Furthermore, we distill lessons for next-generation temporal benchmarks, arguing that Web3 systems provide a high-fidelity sandbox for studying temporal challenges, such as churn, risk, and evolution that are fundamental to the wider Web.
