Heuristic Classification of Thoughts Prompting (HCoT): Integrating Expert System Heuristics for Structured Reasoning into Large Language Models
Summary: arXiv:2604.12390v1 Announce Type: new
Abstract
This paper addresses two limitations of large language models (LLMs) in solving complex problems: (1) their reasoning processes exhibit Bayesian-like stochastic generation, where each token is sampled from a context-dependent probability distribution, leading to inherently random decision trajectories rather than deterministic planning; (2) the reasoning and decision-making mechanisms are statically decoupled, meaning dynamically retrieved domain knowledge fails to dynamically adjust the underlying reasoning strategy. These dual deficiencies result in initial decisions lacking strategic anchoring and reasoning chains often failing to converge on correct solutions, as stochastic generation lacks mechanisms for trajectory correction or knowledge-guided optimization during sequential reasoning.
Proposed Solution
To resolve these issues, we propose a problem-solving method integrated into the LLM’s generation process to guide reasoning. This method, compatible with numerous LLMs and featuring reusable solutions, is grounded in a novel Heuristic-Classification-of-Thoughts prompting schema (HCoT). HCoT synergizes the LLM’s reasoning ability with a structured problem space via a heuristic classification model that controls the reasoning process and provides reusable abstract solutions.
Evaluation and Results
Evaluated on two complex inductive reasoning tasks with ill-defined search spaces, HCoT outperforms existing approaches (e.g., Tree-of-Thoughts and Chain-of-Thoughts prompting) in performance. On the well-structured 24 Game task, HCoT demonstrates significantly higher token efficiency compared to the state-of-the-art Tree-of-Thoughts-Breadth-First-Search.
Key Advantages
HCoT achieves notable improvements in the following areas:
- Accuracy: HCoT consistently converges on correct solutions, demonstrating enhanced reasoning capabilities.
- Token Efficiency: By optimizing the token usage, HCoT maintains high performance while reducing computational costs.
- Pareto Frontier Balance: HCoT achieves a strong trade-off between performance and computational cost, ensuring efficiency in processing.
Conclusion
The Heuristic Classification of Thoughts prompting schema presents a significant advancement in the field of large language models. By integrating structured reasoning and expert system heuristics, HCoT addresses critical challenges faced by existing LLMs, paving the way for more effective problem-solving approaches in artificial intelligence.