$\text{C}^2\text{P}$: Featuring Large Language Models with Causal Reasoning

TL;DR

The paper introduces Causal Chain of Prompting (C^2P), a novel framework that enhances large language models' causal reasoning abilities without external tools, significantly improving their accuracy on synthetic and real-world datasets.

Contribution

C^2P is the first autonomous framework for causal reasoning in LLMs, improving reasoning accuracy by over 30% and demonstrating effectiveness in diverse real-world scenarios.

Findings

Reasoning accuracy improved by over 30% on synthetic datasets.

Few-shot learning with C^2P increased accuracy by over 20%.

Enhanced reasoning performance in real-world story datasets.

Abstract

Causal reasoning is one of the primary bottlenecks that Large Language Models (LLMs) must overcome to attain human-level intelligence. Recent studies indicate that LLMs display near-random performance on reasoning tasks. To address this, we introduce the Causal Chain of Prompting ($\text{C}^2\text{P}$), a reasoning framework that aims to equip current LLMs with causal reasoning capabilities as the first framework of its kind operating autonomously without relying on external tools or modules during both the causal learning and reasoning phases. To evaluate the performance of $\text{C}^2\text{P}$, we first demonstrate that reasoning accuracy improved by over $30.7\%$ and $25.9\%$ for GPT-4 Turbo and LLaMA 3.1, respectively, when using our framework, compared to the same models without $\text{C}^2\text{P}$ on a synthetic benchmark dataset. Then, using few-shot learning of the same LLMs with $\text{C}^2\text{P}$, the reasoning accuracy increased by more than $20.05\%$ and $20.89\%$, respectively, with as few as ten examples, compared to the corresponding LLMs without $\text{C}^2\text{P}$ on the same dataset. To evaluate $\text{C}^2\text{P}$ in realistic scenarios, we utilized another benchmark dataset containing natural stories across various fields, including healthcare, medicine, economics, education, social sciences, environmental science, and marketing. The results show improved reasoning when $\text{C}^2\text{P}$ is applied, compared to cases where our framework is not used, which often leads to random and hallucinated responses. By showing the improved performance of few-shot learned GPT-4 Turbo and LLaMA 3.1 with $\text{C}^2\text{P}$, we demonstrate the generalizability of our framework.