tattoo-tool-grounded-thinking-prm

Build process reward models for tabular reasoning by grounding intermediate reasoning steps in executable tool operations. Train on 60K+ step-level annotations combining verification rationales with tool execution feedback. TaTToo achieves 30.9% improvement over policies using only final rewards, outperforming 72B reasoning models with just 8B parameters vi…

Independent GitHub snapshot · Automated classification · Original documentation governs