Paper: arXiv 2507.20494

Authors: Dhanashekar Kandaswamy, Ashutosh Sahoo, Akshay SP, Gurukiran S, Parag Paul, Girish G N

Abstract

As decentralized finance (DeFi) evolves, distinguishing between user behaviors - liquidity provision versus active trading - has become vital for risk modeling and on-chain reputation. We propose a behavioral scoring framework for Uniswap that assigns two complementary scores: a Liquidity Provision Score that assesses strategic liquidity contributions, and a Swap Behavior Score that reflects trading intent, volatility exposure, and discipline. The scores are constructed using rule-based blueprints that decompose behavior into volume, frequency, holding time, and withdrawal patterns. To handle edge cases and learn feature interactions, we introduce a deep residual neural network with densely connected skip blocks inspired by the U-Net architecture. We also incorporate pool-level context such as total value locked (TVL), fee tiers, and pool size, allowing the system to differentiate similar user behaviors across pools with varying characteristics. Our framework enables context-aware and scalable DeFi user scoring, supporting improved risk assessment and incentive design. Experiments on Uniswap v3 data show its usefulness for user segmentation and protocol-aligned reputation systems. Although we refer to our metric as zScore, it is independently developed and methodologically different from the cross-protocol system proposed by Udupi et al. Our focus is on role-specific behavioral modeling within Uniswap using blueprint logic and supervised learning.

Complexity vs Empirical Score

  • Math Complexity: 5.5/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a hybrid behavioral scoring mechanism for Uniswap v3 using a U-Net-inspired residual neural network combined with heuristic blueprints. It strikes a good balance between mathematical formulation of neural networks and empirical application to on-chain transaction data, though exact code and dataset availability are limited.

Research Flowchart

  flowchart TD
    A[Research Goal: Differentiate & score DeFi wallet behaviors for on-chain reputation] --> B[Input Data: Uniswap v3 transactions & Pool-level Context TVL, Fee Tiers]
    B --> C[Methodology: Define Blueprint Rules for Volume, Frequency, & Holding Time]
    C --> D[Compute Initial Scores: Liquidity Provision Score & Swap Behavior Score]
    D --> E[Train Deep ResNet with U-Net Skip Blocks to learn feature interactions]
    E --> F[Outcomes: Context-aware zScore for wallet segmentation & risk assessment]

Cited References (24)

  1. Evaluating crypto exchanges in the absence of governmental frameworks-a multiple criteria scoring model (2019)
  2. Crypto exchanges and credit risk: Modeling and forecasting the probability of closure (2021)
  3. Review of machine learning models for credit scoring analysis (2020)
  4. RiskProp: Account risk rating on Ethereum via de-anonymous score and network propagation (2024)
  5. zScore: A universal decentralised reputation system for the blockchain economy (2025) — link
  6. A Reputation Scoring Framework for Lending Protocols Using the PageRank Algorithm (2024)
  7. Risk Forecasting Comparisons in Decentralized Finance: An Approach in Constant Product Market Makers (2025)
  8. Decentralized exchange (2017)
and 16 more
  1. Uniswap and the Emergence of the Decentralized Exchange (2021)
  2. Decentralized exchanges: The “wild west” of cryptocurrency trading (2021)
  3. Centralized exchanges vs. decentralized exchanges in cryptocurrency markets: A systematic literature review (2024)
  4. Risks and returns of uniswap v3 liquidity providers (2022)
  5. Uniswap: Impermanent loss and risk profile of a liquidity provider (2021) — link
  6. Decentralized credit scoring: Black box 3.0 (2024)
  7. Credit Score-Based Lending System On the Ethereum Platform (2024)
  8. Wire: Web3 integrated reputation engine (2024)
  9. Statistical and machine learning models in credit scoring: A systematic literature survey (2020)
  10. Machine learning for credit scoring: Improving logistic regression with non-linear decision-tree effects (2022)
  11. A study on credit scoring modeling with different feature selection and machine learning approaches (2020)
  12. Machine learning approach for credit scoring (2020) — link
  13. DeFi protocol risks: The paradox of DeFi (2021)
  14. DeFi risks and the decentralisation illusion (2021)
  15. A machine learning approach for micro-credit scoring (2021)
  16. Transparency, auditability, and explainability of machine learning models in credit scoring (2022)

0 of 24 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.