A simple yet effective post-processing method for detecting unknown intent in dialogue systems based on pre-trained deep neural network classifiers

Code for submitted paper

Autonomous AI prediction market bot for Polymarket. Claude analyzes markets without seeing the price, calibrates its own confidence, sizes with Kelly criterion. Paper mode by default.

🔮 Retrieval-augmented forecasting agent for Prophet Arena — calibrated probability predictions with cost-tiered LLM routing

Confidence calibration toolkit for LLM verbalized-probability outputs. Real benchmark on 998 BoolQ questions with Llama-3.1-8B: ECE 0.148 -> 0.030, log-loss 3.9 -> 0.41.

Probability calibration for binary classifiers: Platt scaling, isotonic regression, ECE, reliability diagram.