--- title: Crypto & DeFi url: https://www.ml-quant.com/topics/crypto-defi/ site: ML-Quant (https://www.ml-quant.com) updated: 2026-09-26 license: Summaries CC BY 4.0; links go to the original sources index: https://www.ml-quant.com/llms.txt --- # Crypto & DeFi Crypto assets, DeFi, stablecoins and blockchain markets. 294 papers featured; 13 in the last 12 months. Papers featured per quarter: 2023 Q2 13, 2023 Q3 29, 2023 Q4 31, 2024 Q1 35, 2024 Q2 37, 2024 Q3 33, 2024 Q4 14, 2025 Q1 40, 2025 Q2 37, 2025 Q3 12, 2025 Q4 11, 2026 Q1 0, 2026 Q2 0, 2026 Q3 2 ## Most cited - [Factors Influencing Cryptocurrency Prices: Evidence from Bitcoin, Ethereum, Dash, Litcoin, and Monero](https://www.ml-quant.com/papers/arxiv/2511.22782/): 225 citations. The study examines the price factors influencing five major cryptocurrencies from 2010-2018, highlighting the roles of market conditions, long-term appeal, and the SP500 index. - [Decentralised Finance and Automated Market Making: Execution and Speculation](https://www.ml-quant.com/papers/ssrn/4144743/): 49 citations. The study investigates automated market makers, particularly constant product market makers, and develops two optimal trading strategies using stochastic optimal control tools, as demonstrated with Uniswap v3 data. - [Decentralised Finance and Automated Market Making: Predictable Loss and Optimal Liquidity Provision](https://www.ml-quant.com/papers/arxiv/2309.08431/): 47 citations. The study outlines the wealth dynamics of strategic liquidity providers in constant product markets, proposes an optimal liquidity provision strategy, and uses Uniswap v3 data to show its effectiveness. - [The Influence of ChatGPT on Artificial Intelligence Related Crypto Assets: Evidence from a Synthetic Control Analysis](https://www.ml-quant.com/papers/doi/10-1016-j-frl-2023-103993/): 37 citations. New method proposed to compute security prices without third party involvement. - [Price Discovery in Cryptocurrency Markets](https://www.ml-quant.com/papers/arxiv/2506.08718/): 29 citations. Centralized markets typically lead in Ethereum price discovery compared to decentralized exchanges, affecting liquidity, arbitrage, and market efficiency. - [The Shape of Money Laundering: Subgraph Representation Learning on the Blockchain with the Elliptic2 Dataset](https://www.ml-quant.com/papers/arxiv/2404.19109/): 28 citations. The article presents Elliptic2, a dataset of Bitcoin subgraphs, to help understand and combat money laundering in cryptocurrency. - [Deep learning and NLP in cryptocurrency forecasting: Integrating financial, blockchain, and social media data](https://www.ml-quant.com/papers/arxiv/2311.14759/): 27 citations. The study uses Machine Learning and Natural Language Processing to predict Bitcoin and Ethereum prices using Twitter and Reddit data, improving forecasting accuracy. - [A Data Engineering Framework for Ethereum Beacon Chain Rewards: From Data Collection to Decentralization Metrics](https://www.ml-quant.com/papers/arxiv/2402.11170/): 25 citations. A study of consensus reward data from the Ethereum Beacon chain offers insights into reward distribution and evolution, aiding in the assessment and refinement of blockchain systems' decentralization, security, and efficiency. - [Axioms for Automated Market Makers: A Mathematical Framework in FinTech and Decentralized Finance](https://www.ml-quant.com/papers/arxiv/2210.01227/): 25 citations. Mathematical Framework: The research introduces a new framework for Automated Market Makers, suggesting a unique fee structure and a novel AMM that reduces divergence loss. - [A comparison of cryptocurrency volatility-benchmarking new and mature asset classes](https://www.ml-quant.com/papers/arxiv/2404.04962/): 24 citations. The paper analyzes the factors influencing cryptocurrency volatility from 2020 to 2022, highlighting that positive market returns, positive signed volatility, and negative daily leverage increase price volatility. - [A dataset of Uniswap daily transaction indices by network](https://www.ml-quant.com/papers/arxiv/2312.02660/): 24 citations. The study explores the effect of Layer 2 solutions on DeFi by analyzing millions of transactions from Uniswap, offering insights into adoption, scalability, and decentralization in the DeFi sector. - [Decentralized Finance: Protocols, Risks, and Governance](https://www.ml-quant.com/papers/arxiv/2312.01018/): 22 citations. Protocols, Risks, Governance: The article discusses the benefits of decentralized finance (DeFi) over traditional finance, the function of smart contracts, and the associated risks, highlighting the need for more research on scalability and auditing. ## Latest - [ETH-TraceBench: A Large-Scale Event-Stream Benchmark for Ethereum DeFi under Temporal, Protocol, and Contract Shift](https://www.ml-quant.com/papers/arxiv/2609.23659/) (2026-09-25): Introduces a large-scale benchmark on 1.35 billion Ethereum transactions to evaluate DeFi representations under temporal, protocol, and contract drift, revealing model degradation on unseen pools. - [Stablecoins Meet the Mundell–Fleming Trilemma](https://www.ml-quant.com/papers/repec/fip-fednsr-103702/) (2026-09-25): Wallet-level stablecoin data shows crisis countries experience inflows during banking restrictions; this endogenizes capital mobility and tightens monetary policy constraints. - [Cash vs. Crypto in DeFi](https://www.ml-quant.com/papers/ssrn/4478395/) (2025-12-28): The article discusses how cryptocurrencies can improve societal functions compared to traditional currencies and emphasizes the innovations needed to build confidence in decentralized finance. - [Bitcoin's Price Alchemy: Unraveling the Influence of Macro Announcements on Volatility and Trading Volume in an Era of Rising Inflation](https://www.ml-quant.com/papers/ssrn/4478184/) (2025-12-28): Bitcoin's price volatility significantly reacts to FOMC and CPI announcements, showing unique patterns during inflation. - [The Impact of Bitcoin ETF Approval on Bitcoin's Hedging Properties Against Traditional Assets](https://www.ml-quant.com/papers/arxiv/2512.12815/) (2025-12-19): Changing Dynamics with Traditional Assets: In January 2024, the approval of a Bitcoin Spot ETF boosted interest from big investors and made Bitcoin more closely linked to stock markets, while its connections to gold remained steady and its relationship with the U.S. Dollar stayed negative. - [Market Effects of Order Flow in Crypto](https://www.ml-quant.com/papers/ssrn/4961610/) (2025-12-01): Analysis indicates that payment for order flow in crypto markets increases trading costs and reduces volumes, especially for assets beyond Bitcoin and Ethereum, after new tokens are introduced. - [DeXposure: A Dataset and Benchmarks for Inter-protocol Credit Exposure in Decentralized Financial Networks](https://www.ml-quant.com/papers/arxiv/2511.22314/) (2025-12-01): Credit Exposure Dataset: The DeXposure dataset offers a comprehensive resource for analyzing credit exposure in decentralized finance, featuring 43.7 million entries for financial machine learning research. - [Factors Influencing Cryptocurrency Prices: Evidence from Bitcoin, Ethereum, Dash, Litcoin, and Monero](https://www.ml-quant.com/papers/arxiv/2511.22782/) (2025-12-01): The study examines the price factors influencing five major cryptocurrencies from 2010-2018, highlighting the roles of market conditions, long-term appeal, and the SP500 index. - [Denoising Complex Covariance Matrices with Hybrid ResNet and Random Matrix Theory: Cryptocurrency Portfolio Applications](https://www.ml-quant.com/papers/arxiv/2510.19130/) (2025-10-27): Builds a robust crypto covariance model using power laws and Random Matrix Theory with ResNet corrections to create stable, profitable portfolios across market regimes. - [Mechanism design and equilibrium analysis of smart contract mediated resource allocation](https://www.ml-quant.com/papers/arxiv/2510.05504/) (2025-10-09): The research introduces a design framework for smart contract-based resource allocation, aiming to balance efficiency and fairness in decentralized systems, validated through tests and a real-world case study. - [Smart Contract Adoption in Derivative Markets under Bounded Risk: An Optimization Approach](https://www.ml-quant.com/papers/arxiv/2510.07006/) (2025-10-09): The research combines structural theory with real-world validation to analyze the adoption of smart contracts under bounded risk, showing that while adoption choices are robust, their financial outcomes are fragile. - [Smart Contract Adoption under Discrete Overdispersed Demand: A Negative Binomial Optimization Perspective](https://www.ml-quant.com/papers/arxiv/2510.05487/) (2025-10-09): The paper presents an optimization framework that merges dynamic Negative Binomial demand modeling with smart contract adoption, offering guidance for balancing inventory costs, service levels, and implementation expenses in high-variance demand scenarios. - [Improving Cryptocurrency Pump-and-Dump Detection through Ensemble-Based Models and Synthetic Oversampling Techniques](https://www.ml-quant.com/papers/arxiv/2510.00836/) (2025-10-03): The study uses SMOTE and advanced ensemble learning models like XGBoost and LightGBM to effectively detect pump and dump manipulation in cryptocurrency markets, enhancing market transparency and stability. - [How Exclusive are Ethereum Transactions? Evidence from non-winning blocks](https://www.ml-quant.com/papers/arxiv/2509.16052/) (2025-09-22): Dominant Source of Builder Revenues: The study shows that exclusive transactions, found only in blocks from one builder, make up 84% of total fees paid by transactions in winning blocks on Ethereum's blockchain. This makes them the main source of builder revenues. - [Banking 2.0: The Stablecoin Banking Revolution - How Digital Assets Are Reshaping Global Finance](https://www.ml-quant.com/papers/arxiv/2508.11395/) (2025-08-20): Stablecoins, combining cryptocurrency and traditional finance, are poised to transform the global financial system by addressing flaws in current fiat currencies and promoting a more globally connected financial system. - [Mapping Microscopic and Systemic Risks in TradFi and DeFi: a literature review](https://www.ml-quant.com/papers/arxiv/2508.12007/) (2025-08-20): The research offers a comparative analysis of systemic risks in traditional and decentralized finance, underlining their distinct features and the idea of 'crosstagion', where instability can spread between the two systems. - [Universal Patterns in the Blockchain: Analysis of EOAs and Smart Contracts in ERC20 Token Networks](https://www.ml-quant.com/papers/arxiv/2508.04671/) (2025-08-12): A study of 44 million Ethereum transactions reveals distinct patterns, showing differences between human and automated transaction behaviors in blockchain ecosystems. - [The Marginal Effects of Ethereum Network MEV Transaction Re-Ordering](https://www.ml-quant.com/papers/arxiv/2508.04003/) (2025-08-07): Two MEV builders, capable of manipulating transaction orders, now generate almost 80% of Ethereum blocks. This has led to frequent 'sandwich attacks' and indicates a need for changes like gas fee priority or private transaction pools. - [Mitigating Financial Frictions in Agriculture: A Framework for Stablecoin Adoption](https://www.ml-quant.com/papers/arxiv/2507.14970/) (2025-07-25): Stablecoins Solution: Fiat-collateralized stablecoins, a type of digital currency, could help solve financial issues in the global agricultural sector by reducing trade costs, increasing supply chain efficiency, and broadening credit access. - [A Comparative Analysis of Statistical and Machine Learning Models for Outlier Detection in Bitcoin Limit Order Books](https://www.ml-quant.com/papers/arxiv/2507.14960/) (2025-07-25): Research comparing statistical methods and machine learning for anomaly detection in cryptocurrency limit order books shows the Empirical Covariance model is the most effective, beating a standard Buy-and-Hold benchmark by 6.70%. - [A New Incentive Model For Content Trust](https://www.ml-quant.com/papers/arxiv/2507.09972/) (2025-07-17): The paper proposes a decentralized, incentive-based method for verifying digital content authenticity using smart contracts and digital identity to fight misinformation. - [Quantifying Crypto Portfolio Risk: A Simulation-Based Framework Integrating Volatility, Hedging, Contagion, and Monte Carlo Modeling](https://www.ml-quant.com/papers/arxiv/2507.08915/) (2025-07-17): The paper introduces a modular simulation framework for analyzing cryptocurrency portfolio risk, incorporating various testing and modeling methods, and validated with recent cryptocurrency data. - [The Feasibility of MBSs as Decentralized Autonomous Organizations](https://www.ml-quant.com/papers/ssrn/4761692/) (2025-07-10): The article investigates the use of modern fintech tools such as asset tokenization, smart contracts, and decentralized autonomous organizations for structuring mortgage-backed security contracts, which could enable real-time ratings systems and enhance market efficiency. - [Pathwise Roughness of Bitcoin Realized Volatility: Stability Across Time, Sampling, and Volatility Measures](https://www.ml-quant.com/papers/arxiv/2507.00575/) (2025-07-03): Rough volatility models are found to be misaligned with Bitcoin volatility due to a multifractal structure that contradicts the homogeneity assumptions of rough volatility estimation. - [From Means to Medians: Optimal Benchmark Design](https://www.ml-quant.com/papers/arxiv/2506.22142/) (2025-07-03): A study using a model inspired by blockchain smart contracts reveals that the best price benchmark varies based on the significance of fixed and variable manipulation costs. - [The Autonomy of the Lightning Network: A Mathematical and Economic Proof of Structural Decoupling from BTC](https://www.ml-quant.com/papers/arxiv/2506.19333/) (2025-06-25): Synthetic Financial System: The study reveals that the Lightning Network, a system designed to speed up Bitcoin transactions, deviates from Bitcoin's original model. It shows increased transaction fees due to capacity limitations and suggests that it resembles a shadow banking system, lacking transparency and guaranteed settlements. - [PRICING OPTIONS ON THE CRYPTOCURRENCY FUTURES CONTRACTS](https://www.ml-quant.com/papers/arxiv/2506.14614/) (2025-06-18): Research indicates that Kou and Bates models, which include jumps and stochastic volatility, are the most accurate for pricing Bitcoin and Ether cryptocurrency options. - [Automated Risk Management Mechanisms in DeFi Lending Protocols: A Crosschain Comparative Analysis of Aave and Compound](https://www.ml-quant.com/papers/arxiv/2506.12855/) (2025-06-18): The latest versions (v3) of Aave and Compound lending protocols show improved risk management compared to their previous versions (v2), with liquidation events boosting total value and revenue, particularly on the L2 blockchain. - [Becoming Immutable: How Ethereum is Made](https://www.ml-quant.com/papers/arxiv/2506.04940/) (2025-06-11): Ethereum blockchain data shows 85% of transaction fees are from exclusive transactions, causing user transaction delays, and two bots are trading more efficiently than Binance. - [Exploring Microstructural Dynamics in Cryptocurrency Limit Order Books: Better Inputs Matter More Than Stacking Another Hidden Layer](https://www.ml-quant.com/papers/arxiv/2506.05764/) (2025-06-11): Simpler models with data preprocessing and hyperparameter tuning can match or surpass complex networks in short-term cryptocurrency price forecasting.