English

Beyond Labels: Aligning Large Language Models with Human-like Reasoning

Computation and Language 2024-08-23 v1 Artificial Intelligence Machine Learning

Abstract

Aligning large language models (LLMs) with a human reasoning approach ensures that LLMs produce morally correct and human-like decisions. Ethical concerns are raised because current models are prone to generating false positives and providing malicious responses. To contribute to this issue, we have curated an ethics dataset named Dataset for Aligning Reasons (DFAR), designed to aid in aligning language models to generate human-like reasons. The dataset comprises statements with ethical-unethical labels and their corresponding reasons. In this study, we employed a unique and novel fine-tuning approach that utilizes ethics labels and their corresponding reasons (L+R), in contrast to the existing fine-tuning approach that only uses labels (L). The original pre-trained versions, the existing fine-tuned versions, and our proposed fine-tuned versions of LLMs were then evaluated on an ethical-unethical classification task and a reason-generation task. Our proposed fine-tuning strategy notably outperforms the others in both tasks, achieving significantly higher accuracy scores in the classification task and lower misalignment rates in the reason-generation task. The increase in classification accuracies and decrease in misalignment rates indicate that the L+R fine-tuned models align more with human ethics. Hence, this study illustrates that injecting reasons has substantially improved the alignment of LLMs, resulting in more human-like responses. We have made the DFAR dataset and corresponding codes publicly available at https://github.com/apurba-nsu-rnd-lab/DFAR.

Keywords

Cite

@article{arxiv.2408.11879,
  title  = {Beyond Labels: Aligning Large Language Models with Human-like Reasoning},
  author = {Muhammad Rafsan Kabir and Rafeed Mohammad Sultan and Ihsanul Haque Asif and Jawad Ibn Ahad and Fuad Rahman and Mohammad Ruhul Amin and Nabeel Mohammed and Shafin Rahman},
  journal= {arXiv preprint arXiv:2408.11879},
  year   = {2024}
}

Comments

Accepted in ICPR 2024

R2 v1 2026-06-28T18:19:55.462Z