IDEAS home Printed from https://ideas.repec.org/p/arx/papers/2606.19501.html

DeXposure-Claw: An Agentic System for DeFi Risk Supervision

Author

Listed:
  • Aijie Shu
  • Bowei Chen
  • Wenbin Wu
  • Cathy Yi-Hsuan Chen
  • Fengxiang He

Abstract

Decentralized finance exposes supervisors to fast-moving, networked credit risks. General-purpose LLM agents fit this setting poorly: they over-read weak evidence and recommend high-stakes interventions, while existing evaluations offer no regulator-aligned way to measure the resulting false alarms. We introduce DeXposure-Claw, a forecast-grounded agentic supervision system that routes LLM decisions through structured evidence: (1) DeXposure-FM, a graph time-series foundation model, forecasts future exposure networks; (2) deterministic monitors and stress scenarios then turn those forecasts into typed alerts, attribution signals, and scenario evidence; and (3) data-health and confidence gates constrain escalation before DeXposure-Claw emits auditable supervisory tickets with rationales. We further develop DeXposure-Bench, a six-axis evaluation harness, whose decision axis scores tickets against a regulator-aligned absolute-loss ground truth and an explicit false-intervention rate. Experiments on five years of weekly real data fully support our system. Code is at https://github.com/EVIEHub/DeXposure-Claw.

Suggested Citation

  • Aijie Shu & Bowei Chen & Wenbin Wu & Cathy Yi-Hsuan Chen & Fengxiang He, 2026. "DeXposure-Claw: An Agentic System for DeFi Risk Supervision," Papers 2606.19501, arXiv.org, revised Jun 2026.
    • Handle: RePEc:arx:papers:2606.19501
    as

    Download full text from publisher

    File URL: https://arxiv.org/pdf/2606.19501
    File Function: Latest version
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Qian'ang Mao & Yuxuan Zhang & Jiaman Chen & Wenjun Zhou & Jiaqi Yan, 2025. "Know Your Intent: An Autonomous Multi-Perspective LLM Agent Framework for DeFi User Transaction Intent Mining," Papers 2511.15456, arXiv.org.
    2. Lukas Gonon & Thilo Meyer-Brandis & Niklas Weber, 2024. "Computing Systemic Risk Measures with Graph Neural Networks," Papers 2410.07222, arXiv.org, revised Oct 2025.
    3. Sun, Yiqiao & de Bondt, Gabe, 2025. "Enhancing GDP nowcasts with ChatGPT: a novel application of PMI news releases," Working Paper Series 3063, European Central Bank.
    4. Byeungchun Kwon & Taejin Park & Fernando Perez-Cruz & Phurichai Rungcharoenkitkul, 2024. "Large language models: a primer for economists," BIS Quarterly Review, Bank for International Settlements, December.
    5. Andrea Carriero & Davide Pettenuzzo & Shubhranshu Shekhar, 2024. "Macroeconomic Forecasting with Large Language Models," Papers 2407.00890, arXiv.org, revised Sep 2025.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Aijie Shu & Wenbin Wu & Gbenga Ibikunle & Fengxiang He, 2026. "DeXposure-FM: A Time-series, Graph Foundation Model for Credit Exposures and Stability on Decentralized Financial Networks," Papers 2602.03981, arXiv.org, revised Jun 2026.
    2. Umberto Collodel, 2025. "Interpreting the Interpreter: Can We Model post-ECB Conferences Volatility with LLM Agents?," Papers 2508.13635, arXiv.org, revised May 2026.
    3. Kwon, Byeungchun & Park, Taejin & Rungcharoenkitkul, Phurichai & Smets, Frank, 2025. "Parsing the Pulse: Decomposing Macroeconomic Sentiment with LLMs," CEPR Discussion Papers 20828, Centre for Economic Policy Research.
    4. Donggyu Lee & Hyeok Yun & Jungwon Kim & Junsik Min & Sungwon Park & Sangyoon Park & Jihee Kim, 2026. "Ideological Bias in LLMs' Economic Causal Reasoning," Papers 2604.21334, arXiv.org, revised Jul 2026.
    5. Douglas Araujo & Rafael Schmidt & Olivier Sirello & Bruno Tissot & Ricardo Villarreal, 2025. "Governance and implementation of artificial intelligence in central banks," IFC Reports 18, Bank for International Settlements.
    6. Yichen Luo & Yebo Feng & Jiahua Xu & Yang Liu, 2026. "Resisting Manipulative Bots in Meme Coin Copy Trading: A Multi-Agent Approach with Chain-of-Thought Reasoning," Papers 2601.08641, arXiv.org, revised Feb 2026.
    7. Camille Jehle & Florian Le Gallo, 2025. "Europe in the Headlines: What Two Decades of French News Reveal about EU Sentiment," Working papers 1008, Banque de France.
    8. George Fatouros & Kostas Metaxas & John Soldatos & Manos Karathanassis, 2025. "MarketSenseAI 2.0: Enhancing Stock Analysis through LLM Agents," Papers 2502.00415, arXiv.org, revised Oct 2025.
    9. SEKINE, Toshitaka & WADA, Tetsuro, 2025. "How Did People Tweet against Inflation in Japan?," Discussion paper series HIAS-E-150, Hitotsubashi Institute for Advanced Study, Hitotsubashi University.
    10. Hui Gong, 2026. "Agent-to-Agent Finance: Blockchain Payments and Trust Infrastructure for Autonomous AI Agents," Papers 2607.00245, arXiv.org.
    11. Andrea Carriero & Davide Pettenuzzo & Shubhranshu Shekhar, 2026. "MACROCAST: A Vintage-Consistent Time Series Foundation Model for Real-Time Macroeconomic Forecasting," Papers 2606.28670, arXiv.org.
    12. Batuhan Koyuncu & Byeungchun Kwon & Marco Jacopo Lombardi & Fernando Perez-Cruz & Hyun Song Shin, 2026. "BISTRO: a general purpose oracle for macroeconomic time series," BIS Quarterly Review, Bank for International Settlements, March.
    13. Ana Lorena Jiménez-Preciado & Mario Alejandro Durán-Saldivar & Salvador Cruz-Aké & Francisco Venegas-Martínez, 2025. "Hawkish or Dovish? That Is the Question: Agentic Retrieval of FED Monetary Policy Report," Mathematics, MDPI, vol. 13(20), pages 1-23, October.
    14. Batuhan Koyuncu & Byeungchun Kwon & Marco Jacopo Lombardi & Fernando Perez-Cruz & Hyun Song Shin, 2026. "Introducing BISTRO: a foundational model for unconditional and conditional forecasting of macroeconomic time series," BIS Working Papers 1337, Bank for International Settlements.
    15. Philippe Goulet Coulombe, 2025. "Ordinary Least Squares as an Attention Mechanism," Papers 2504.09663, arXiv.org, revised Jan 2026.

    More about this item

    Statistics

    Access and download statistics

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:arx:papers:2606.19501. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: arXiv administrators (email available below). General contact details of provider: https://arxiv.org/ .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.