Track and Trace: Automatically Uncovering Cross-chain Transactions in the Multi-blockchain Ecosystems

TL;DR

Cross-chain DeFi bridges create complex, multi-ledger security challenges for tracing asset flows. The authors introduce ABCTracer, an automated, bi-directional cross-chain transaction tracing framework that combines cross-chain semantic extraction, explicit clue learning via named entity recognition, and implicit clue encoding via information retrieval to pair source deposits with destination withdrawals. On a real-world dataset of 12 mainstream bridges, ABCTracer achieves forward, backward, and bidirectional tracing $F_1$ scores of $94.92\%$, $89.58\%$, and $91.75\%$, respectively, and demonstrates robust automatic learning and generalization to unseen bridges. The approach is further validated by applying ABCTracer to cross-chain attack and money-laundering cases, identifying multiple notable transactions and revealing patterns that enhance DeFi security. Overall, ABCTracer advances cross-chain tracing beyond prior rule-based and CeFi-centric methods, enabling scalable AML and threat-detection capabilities across multi-chain ecosystems.

Abstract

Cross-chain technology enables seamless asset transfer and message-passing within decentralized finance (DeFi) ecosystems, facilitating multi-chain coexistence in the current blockchain environment. However, this development also raises security concerns, as malicious actors exploit cross-chain asset flows to conceal the provenance and destination of assets, thereby facilitating illegal activities such as money laundering. Consequently, the need for cross-chain transaction traceability has become increasingly urgent. Prior research on transaction traceability has predominantly focused on single-chain and centralized finance (CeFi) cross-chain scenarios, overlooking DeFispecific considerations. This paper proposes ABCTRACER, an automated, bi-directional cross-chain transaction tracing tool, specifically designed for DeFi ecosystems. By harnessing transaction event log mining and named entity recognition techniques, ABCTRACER automatically extracts explicit cross-chain cues. These cues are then combined with information retrieval techniques to encode implicit cues. ABCTRACER facilitates the autonomous learning of latent associated information and achieves bidirectional, generalized cross-chain transaction tracing. Our experiments on 12 mainstream cross-chain bridges demonstrate that ABCTRACER attains 91.75% bi-directional traceability (F1 metrics) with self-adaptive capability. Furthermore, we apply ABCTRACER to real-world cross-chain attack transactions and money laundering traceability, thereby bolstering the traceability and blockchain ecological security of DeFi bridging applications.

Figures (10)

• Figure 1: Overview of ABCTracer, including three main modules (M1: cross-chain transaction identification based on semantic extraction, M2: candidate transactions localization based on explicit clues, M3: cross-chain transaction association based on implicit clues).
• Figure 2: Cross-chain transaction identification model that extracts and integrates asset transfer and message-passing semantics, followed by a classifier to distinguish cross-chain from non-cross-chain transactions.
• Figure 3: Example event declaration for withdrawal transaction.
• Figure 4: Overview of the candidate transactions localization model. Input sequences are processed through a pre-trained model to generate embeddings, which are then used in a BiLSTM for contextual learning and passed to a CRF for final token tagging.
• Figure 5: An example of an Allbridge cross-chain transaction from Ethereum to Binance Smart Chain, with implicit cues including the recipient, nonce, and messenger parameters.