Boosting Logical Reasoning in Large Language Models through a New Framework: The Graph of Thought
This addresses the challenge of enhancing logical reasoning in AI models for applications requiring complex problem-solving, representing a novel method rather than an incremental improvement.
The paper tackled the problem of low accuracy in large language models for multi-step logical reasoning by introducing the Graph of Thought (GoT) prompting technique, which achieved accuracy improvements of up to 89.7% over GPT-4 and up to 24% over the state-of-the-art Tree of Thought method on tasks like the 24-point game and polynomial equations.
Recent advancements in large-scale models, such as GPT-4, have showcased remarkable capabilities in addressing standard queries. However, when facing complex problems that require multi-step logical reasoning, their accuracy dramatically decreases. Current research has explored the realm of \textit{prompting engineering} to bolster the inferential capacities of these models. Our paper unveils a pioneering prompting technique, dubbed \textit{Graph of Thoughts (GoT)}. Through testing on a trio of escalating challenges: the 24-point game, resolution of high-degree polynomial equations, and derivation of formulas for recursive sequences, our method outperformed GPT-4, achieving accuracy improvements of $89.7\%$, $86\%$, and $56\%$ for each respective task. Moreover, when juxtaposed with the state-of-the-art (SOTA) prompting method, \textit{Tree of Thought (ToT)}, our approach registered an average accuracy boost of $23\%$, $24\%$, and $15\%$.