Rating
1813
Battle Count: 50
Relevance
9/10
Highly relevant for quantitative traders operating in prediction markets or those interested in cross-asset arbitrage. It provides a rigorous framework for identifying and executing complex multi-leg arbitrage strategies, quantifying the limits of arbitrage due to information discovery costs, and demonstrating the practical application of LLMs in financial contract analysis.
Implementation Complexity
9/10
High complexity. Requires handling massive datasets (billions of order book messages), implementing integer programming solvers for combinatorial optimization, integrating multiple LLMs for semantic relation extraction with dual-audit mechanisms, and simulating execution against full-depth limit order books.
Reproducibility
3/5
The paper provides extensive details on data collection (Polymarket API, on-chain fills), relation extraction prompts, and the integer programming formulation. However, no public code repository URL is explicitly provided in the text, and the raw order book data (73 billion messages) is likely proprietary or too large for standard public distribution, though the methodology is fully described.
Abstract
We develop a framework to identify and quantify arbitrage across logically related prediction-market contracts. We prove that prices outside the convex hull of admissible payoff vectors admit arbitrage and identify optimal arbitrage portfolios through integer programming. Using venue metadata and semantic analysis of resolution criteria, we recover 1.9 million logical relations and compile them into constraint systems covering more than 255,000 contracts. We replay full-depth Polymarket order books to simulate an arbitrageur executing these portfolios against resting orders. Among settled portfolios, guaranteed profit after spreads and fees equals 2.2% of committed capital, while profit at resolution reaches 3.2%. We find that arbitrage opportunities persist longer and attract less observed participation where relation discovery and execution are more costly.
The interactive Everscope explorer (charts, battles, favorites) loads below.