{"ID":2892503,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.14785","arxiv_id":"2507.14785","title":"Exploring the In-Context Learning Capabilities of LLMs for Money Laundering Detection in Financial Graphs","abstract":"The complexity and interconnectivity of entities involved in money laundering demand investigative reasoning over graph-structured data. This paper explores the use of large language models (LLMs) as reasoning engines over localized subgraphs extracted from a financial knowledge graph. We propose a lightweight pipeline that retrieves k-hop neighborhoods around entities of interest, serializes them into structured text, and prompts an LLM via few-shot in-context learning to assess suspiciousness and generate justifications. Using synthetic anti-money laundering (AML) scenarios that reflect common laundering behaviors, we show that LLMs can emulate analyst-style logic, highlight red flags, and provide coherent explanations. While this study is exploratory, it illustrates the potential of LLM-based graph reasoning in AML and lays groundwork for explainable, language-driven financial crime analytics.","short_abstract":"The complexity and interconnectivity of entities involved in money laundering demand investigative reasoning over graph-structured data. This paper explores the use of large language models (LLMs) as reasoning engines over localized subgraphs extracted from a financial knowledge graph. We propose a lightweight pipeline...","url_abs":"https://arxiv.org/abs/2507.14785","url_pdf":"https://arxiv.org/pdf/2507.14785v2","authors":"[\"Erfan Pirmorad\"]","published":"2025-07-20T02:00:21Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\",\"LoRA\"]","has_code":false}