Human-in-the-Loop through Chain-of-Thought

TL;DR

This paper introduces a human-in-the-loop system with Chain-of-Thought prompting and a cost-utility model to enhance reasoning performance of language models while balancing costs, validated through experiments on multiple datasets.

Contribution

It proposes the Manual Correction System (MCS) for improving reasoning via human corrections and introduces CAMLOP, a model to analyze and balance utility and cost in human-in-the-loop systems.

Findings

MCS significantly improves reasoning accuracy over baselines.

CAMLOP effectively balances utility and cost in human-in-the-loop systems.

Experiments on twelve datasets validate the approach's effectiveness.

Abstract

While the emergence of powerful language models along with Chain-of-thought prompting has made automation more and more omnipresent, it sometimes demonstrates its weakness in long-term or multi-step logical reasoning. For example, users don't always get desirable answers for complex mathematical problems without human involvement. Against this background, we present the Manual Correction System (MCS) -- a human-in-the-loop system enhanced by Chain-of-Thought prompting, which explores how manual correction of sub-logics in rationales can improve LLM's reasoning performance. Moving one step forward, considering a system with human-in-the-loop involves more than having humans improve performance but also controlling the cost. Therefore, we post a Cost-utility Analysis Model for Human-in-the-Loop systems (CAMLOP) based on classical economics theory to analyze, quantify and balance the utility and the corresponding cost. We conduct experiments of MCS and CAMLOP with twelve datasets. A significant advantage w.r.t cost and utility proves its superiority over strong baselines.