# Population-Calibrated Graph Screening at 835-Million-Address Scale, with Label-Free Transfer to New Chains

arXiv: https://arxiv.org/abs/2609.03036 · HTML: https://mlsanity.com/paper/2609.03036/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 42 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Builds on (most cited first, 25 of 25)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Graph Convolutional Neural Networks for Web-Scale Recommender Systems](https://mlsanity.com/paper/1806.01973/) (2018)
- [Dota 2 with Large Scale Deep Reinforcement Learning](https://mlsanity.com/paper/1912.06680/) (2019)
- [Heterogeneous Graph Transformer](https://mlsanity.com/paper/2003.01332/) (2020)
- [Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks](https://mlsanity.com/paper/1905.07953/) (2019)
- [Deep Recurrent Q-Learning for Partially Observable MDPs](https://mlsanity.com/paper/1507.06527/) (2015)
- [EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphs](https://mlsanity.com/paper/1902.10191/) (2019)
- [Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics](https://mlsanity.com/paper/1908.02591/) (2019)
- [TESSERACT: Eliminating Experimental Bias in Malware Classification across Space and Time](https://mlsanity.com/paper/1807.07838/) (2018)
- [Dos and Don'ts of Machine Learning in Computer Security](https://mlsanity.com/paper/2010.09470/) (2020)
- ["Real Attackers Don't Compute Gradients": Bridging the Gap Between Adversarial ML Research and Practice](https://mlsanity.com/paper/2212.14315/) (2022)
- [Realistic Synthetic Financial Transactions for Anti-Money Laundering Models](https://mlsanity.com/paper/2306.16424/) (2023)
- [Who Are the Phishers? Phishing Scam Detection on Ethereum via Network Embedding](https://mlsanity.com/paper/1911.09259/) (2019)
- [Scalable Graph Learning for Anti-Money Laundering: A First Look](https://mlsanity.com/paper/1812.00076/) (2018)
- [Provably Powerful Graph Neural Networks for Directed Multigraphs](https://mlsanity.com/paper/2306.11586/) (2023)
- [Demystifying Fraudulent Transactions and Illicit Nodes in the Bitcoin Network for Financial Forensics](https://mlsanity.com/paper/2306.06108/) (2023)
- [TTAGN: Temporal Transaction Aggregation Graph Network for Ethereum Phishing Scams Detection](https://mlsanity.com/paper/2204.13442/) (2022)
- [Quantifying Blockchain Extractable Value: How dark is the forest?](https://mlsanity.com/paper/2101.05511/) (2021)
- [Inspection-L: Self-Supervised GNN Node Embeddings for Money Laundering Detection in Bitcoin](https://mlsanity.com/paper/2203.10465/) (2022)
- [The Shape of Money Laundering: Subgraph Representation Learning on the Blockchain with the Elliptic2 Dataset](https://mlsanity.com/paper/2404.19109/) (2024)
- [Tracing Transactions Across Cryptocurrency Ledgers](https://mlsanity.com/paper/1810.12786/) (2018)
- [On How Zero-Knowledge Proof Blockchain Mixers Improve, and Worsen User Privacy](https://mlsanity.com/paper/2201.09035/) (2022)

## Abstract (arXiv)

Compliance screening of blockchain addresses is, in practice, a lookup against sanctions registries plus clustering heuristics; it fails on unlabelled addresses and on chains with no label coverage at all. We describe a deployed system that scores an address by its position in a multi-chain transaction graph rather than by its presence in a list. The substrate is a single graph of 835,330,427 addresses and 15,826,261,934 edges across five EVM chains; a shared inductive encoder with per-chain normalisation feeds two scoring heads. Decision thresholds are exact quantiles of the score distribution over the full population, scanned per chain segment, so the alert volume is known in advance. We report: label-free transfer: heads trained on two chains recall 0.8598 / 0.8182 / 0.9967 of held-out positives on Base, Arbitrum and Gnosis at a 10-3 population alert rate, with no target-chain labels in head training; a static lead-time replay over 68 external registry events: 40 of 68 (58.8%) flagged at the 0.1% budget, ×152 over an event-level random-flagging baseline, with first on-chain appearance a median of 528.8 days (Ethereum) / 647.8 days (Tron) before public designation; a serving path whose score is bit-identical to the offline artefact at end-to-end p50 151 ms, gated by a 2,882-address drift panel; and an adversarial harness of eight recurrent reinforcement-learned archetypes that passes an 8-criterion degeneracy audit and, on a detector-independent snapshot, exposes a measured blind spot of the deployed heads against synthesised behaviour.