Benchmarking Temporal Web3 Intelligence: Lessons from the FinSurvival 2025 Challenge
This benchmark addresses the lack of shared, reproducible benchmarks for temporal Web3 analytics, which hinders methodological progress and limits technique transfer between Web3 and broader Web domains.
This paper introduces the FinSurvival Challenge 2025, a benchmark for temporal Web3 intelligence, using 21.8 million transaction records from the Aave v3 protocol. The challenge involved 16 survival prediction tasks, demonstrating that domain-aware temporal feature construction significantly outperformed generic modeling approaches.
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.