Does Reward Training Teach AI New Reasoning? A New Paper Says No. Checked and Re-Run
9 November, 2026
By Evan VegaA MOSS @ COLM 2026 workshop paper (arXiv:2611.02574) argues that reinforcement learning with verifiable rewards only sharpens what a model already knows: tracking OLMo 2 1B across 13 training snapshots,..
Read More