Pranav Achar

Where Multi-Step Reasoning Breaks: Measuring Error Propagation and Structured Scaffolds in Small Language Models

Chain-of-thought prompting lets a language model write out its intermediate reasoning, and a wrong intermediate claim can carry through to a wrong final answer. This preprint measures that error propagation directly for Qwen2.5-1.5B, Qwen2.5-3B and Llama-3.2-3B Instruct, run locally in 4-bit. On ProofWriter, every grou...