Parallel-Code World Models: 7B Model Predicts Data Races from Source Code at 72.8% Accuracy

arXiv·medium signal

PCWMs train reasoning LLMs to predict tool outcomes (data races, performance profiles) directly from parallel source code. Novel pipeline samples diverse parallel-coding problems, executes them to record outcomes, then synthesizes hindsight reasoning traces. Fine-tuning a 7B model improves race-outcome prediction from 64.3% to 72.8%—practical for parallel code review and CI/CD race detection.