NoVo: Norm Voting off Hallucinations with Attention Heads in Large Language Models

TL;DR

NoVo is a lightweight, inference-only method that leverages attention head norms to significantly improve factual accuracy in large language models across diverse datasets, surpassing current state-of-the-art methods.

Contribution

The paper introduces Norm Voting (NoVo), a novel approach that uses attention head norms for scalable, zero-shot factual accuracy enhancement in LLMs, with minimal computational overhead.

Findings

NoVo surpasses state-of-the-art accuracy on TruthfulQA MC1 by at least 19 points.

NoVo generalizes well to 20 diverse datasets, improving accuracy in over 90% of cases.

Head norm voting enhances LLM interpretability, robustness, and reliability.

Abstract

Hallucinations in Large Language Models (LLMs) remain a major obstacle, particularly in high-stakes applications where factual accuracy is critical. While representation editing and reading methods have made strides in reducing hallucinations, their heavy reliance on specialised tools and training on in-domain samples, makes them difficult to scale and prone to overfitting. This limits their accuracy gains and generalizability to diverse datasets. This paper presents a lightweight method, Norm Voting (NoVo), which harnesses the untapped potential of attention head norms to dramatically enhance factual accuracy in zero-shot multiple-choice questions (MCQs). NoVo begins by automatically selecting truth-correlated head norms with an efficient, inference-only algorithm using only 30 random samples, allowing NoVo to effortlessly scale to diverse datasets. Afterwards, selected head norms are employed in a simple voting algorithm, which yields significant gains in prediction accuracy. On TruthfulQA MC1, NoVo surpasses the current state-of-the-art and all previous methods by an astounding margin -- at least 19 accuracy points. NoVo demonstrates exceptional generalization to 20 diverse datasets, with significant gains in over 90\% of them, far exceeding all current representation editing and reading methods. NoVo also reveals promising gains to finetuning strategies and building textual adversarial defence. NoVo's effectiveness with head norms opens new frontiers in LLM interpretability, robustness and reliability.