Iterative Graph Alignment

TL;DR

The paper introduces Iterative Graph Alignment (IGA), a scalable, annotation-free rule-based alignment method for large language models, improving their ability to follow rules through iterative supervised fine-tuning.

Contribution

IGA is a novel, annotation-free alignment algorithm that uses iterative graph prompting and collaborative fine-tuning to enhance rule adherence in LLMs.

Findings

73.12% alignment improvement in Claude Sonnet 3.5

86.20% alignment improvement in Llama3-8B-Instruct

Outperforms existing rule-based alignment methods

Abstract

By compressing diverse narratives, LLMs go beyond memorization, achieving intelligence by capturing generalizable causal relationships. However, they suffer from local 'representation gaps' due to insufficient training data diversity, limiting their real-world utility, especially in tasks requiring strict alignment to rules. Traditional alignment methods relying on heavy human annotations are inefficient and unscalable. Recent self-alignment techniques also fall short, as they often depend on self-selection based prompting and memorization-based learning. To address these issues, we introduce Iterative Graph Alignment (IGA), an annotation-free rule-based alignment algorithm. A teacher model (VLM) employs Iterative Graph Prompting (IGP) to create logical graphs and reference answers. The student model (LLM) identifies local knowledge gaps by attempting to align its responses with these references, collaborating with helper models to generate diverse answers. These aligned responses are then used for iterative supervised fine-tuning (SFT). Our evaluations across five rule-based scenarios demonstrate IGP's effectiveness, with a 73.12\% alignment improvement in Claude Sonnet 3.5, and Llama3-8B-Instruct achieving an 86.20\% improvement, outperforming Claude Sonnet 3.5 in rule-based alignment.