Evaluating and Improving Chain-of-Thought Reasoning

Ali Dulaimi

A framework for evaluating and enhancing Chain-of-Thought (CoT) reasoning in LLMs, including causal mediation analysis and faithfulness measurement.

Overview

This project provides tools and methods for evaluating and improving Chain-of-Thought reasoning in large language models, with a focus on causal mediation analysis.

Technologies

  • LLMs
  • CoT (Chain-of-Thought)
  • Causal Inference
  • Python