ECCoT: A Framework for Enhancing Effective Cognition via Chain of Thought in Large Language Model
This addresses interpretability issues in LLMs for users relying on AI-generated reasoning, though it is incremental as it builds on existing CoT methods.
The paper tackles the problem of unreliable and non-transparent outputs in Large Language Models by proposing ECCoT, a framework that validates and refines reasoning chains using Chain of Thought prompting, resulting in improved interpretability, reduced biases, and enhanced trustworthiness in decision-making.
In the era of large-scale artificial intelligence, Large Language Models (LLMs) have made significant strides in natural language processing. However, they often lack transparency and generate unreliable outputs, raising concerns about their interpretability. To address this, the Chain of Thought (CoT) prompting method structures reasoning into step-by-step deductions. Yet, not all reasoning chains are valid, and errors can lead to unreliable conclusions. We propose ECCoT, an End-to-End Cognitive Chain of Thought Validation Framework, to evaluate and refine reasoning chains in LLMs. ECCoT integrates the Markov Random Field-Embedded Topic Model (MRF-ETM) for topic-aware CoT generation and Causal Sentence-BERT (CSBert) for causal reasoning alignment. By filtering ineffective chains using structured ordering statistics, ECCoT improves interpretability, reduces biases, and enhances the trustworthiness of LLM-based decision-making. Key contributions include the introduction of ECCoT, MRF-ETM for topic-driven CoT generation, and CSBert for causal reasoning enhancement. Code is released at: https://github.com/erwinmsmith/ECCoT.git.