https://sergeisolovev.com/feed.xmlSergei Solovev — PreprintsSmart contract security, machine learning, decentralized finance.2026-05-26T12:00:00ZSergei Solovevhttps://sergeisolovev.comCC BY 4.0https://sergeisolovev.com/papers/rag-solidity.htmlWhen Retrieval Hurts: An Honest Evaluation of RAG for Solidity Vulnerability Detection2026-05-01T00:00:00Z2026-05-01T00:00:00ZSergei SolovevEmpirical study showing a sample-size sign reversal in naive RAG for Solidity vulnerability detection: +2.0% Macro-F1 at n=100 flips to -2.7% at n=250 on SolidiFI. Argues for bootstrap confidence intervals in any RAG evaluation.https://sergeisolovev.com/papers/ai-yield-vault.htmlAI-Managed ERC-4626 Yield Vault with Multi-Criteria Decision Making: Design, Implementation, and Formal Verification2026-05-01T00:00:00Z2026-05-01T00:00:00ZSergei SolovevAutonomous AI agent managing an ERC-4626 yield vault via MCDM scoring (APY/Risk/Cost/Stability). EIP-712 signed decisions, UUPS-upgradeable, 67 unit tests, 76,800+ invariant calls with zero violations. Sepolia deployment included.https://sergeisolovev.com/papers/lob-mid-price.htmlWhen Less Is More: Domain-Aware Dual-Branch Recurrent Networks for Limit Order Book Mid-Price Prediction2026-03-26T00:00:00Z2026-03-26T00:00:00ZSergei SolovevDual-branch BiGRU-CNN for high-frequency LOB mid-price prediction. Weighted Pearson rho_w = 0.266, outperforming LightGBM on 58.3% of sequences. Documents feature sufficiency and the negative ensemble effect on HF data (Wunder Fund RNN challenge).https://sergeisolovev.com/papers/ml-vuln-detection.htmlMachine Learning-Based Vulnerability Detection in Ethereum Smart Contracts via EVM Bytecode Feature Engineering2026-02-26T00:00:00Z2026-02-26T00:00:00ZSergei SolovevXGBoost + Optuna on 117,091 Slither-labelled Ethereum contracts achieves F1=0.948 using 65 hand-crafted bytecode features in 15 SWC-mapped semantic categories. Binary classification: recall 0.950, MCC 0.832, PR-AUC 0.990.https://sergeisolovev.com/papers/ocr-vs-donut.htmlOCR-Based vs. End-to-End Transformer Pipelines for Receipt Information Extraction: A Comparative Study on SROIE 20192026-02-26T00:00:00Z2026-02-26T00:00:00ZSergei SolovevEasyOCR + heuristic rules vs. Donut end-to-end Transformer on SROIE 2019. Error taxonomy under image degradation typical of messenger-grade distortions (compression, rotation, blur).