Beyond Forecasting: The Belief-to-Trade Layer in Prediction-Market Agents
For researchers in AI evaluation and prediction markets, this work addresses the gap between forecasting accuracy and trading performance, but the results are based on a single controlled replay, limiting generalizability.
The authors propose Raven-Agent, the first autonomous trading agent for prediction markets, which achieves the only positive return and risk-adjusted return among all tested policies on a controlled replay over an archived decision set.
Forecasting future events has attracted growing attention as a testbed for general-purpose AI. A natural way to ground this evaluation is let the models trade in the prediction markets. Trading, however, requires more than forecasting. Moreover, recent benchmarks report a substantial gap between calibrated probability scores and the trading results. We propose Raven-Agent, to the best of our knowledge, the first autonomous trading agent for prediction markets. On a controlled replay over an archived decision set, our architecture achieves the only positive return and the only positive risk-adjusted return among all tested policies. We have released our code in https://github.com/Alchemist-X/predict-raven .