Small Language Models Fine-tuned to Coordinate Larger Language Models improve Complex Reasoning

arXiv:2310.18338v2 Announce Type: replace
Abstract: Large Language Models (LLMs) prompted to generate chain-of-thought (CoT) exhibit impressive reasoning capabilities. Recent attempts at prompt decomposition toward solving complex, multi-step reasoning problems depend on the ability of the LLM to simultaneously decompose and solve the problem. A significant disadvantage is that foundational LLMs are typically not available for fine-tuning, making adaptation computationally prohibitive. We believe (and demonstrate) that problem decomposition and solution generation are distinct capabilites, better addressed in separate modules, than by …

abstract arxiv capabilities cs.ai cs.cl generate language language models large language large language models llm llms prompt reasoning small small language models solve thought type