Anand, Ayush

The Verifier Gap: Negative Scaling and Syntactic-Logical Divergence in Reasoning Distillation

2026 · Zenodo (CERN European Organization for Nuclear Research)

The rapid proliferation of reasoning-distilled large language models (LLMs) relies on the premise that Supervised Fine-Tuning (SFT) on the reasoning traces of Reinforcement Learning (RL) teachers transfers causal verification capabilities to smaller models. In this …