HalluCana: Fixing LLM Hallucination with A Canary Lookahead

TL;DR

HalluCana introduces a proactive method using canary lookahead to detect and correct factual hallucinations in LLMs during long-form generation, significantly improving quality and efficiency.

Contribution

The paper proposes a novel canary lookahead approach leveraging internal factuality signals to detect and correct hallucinations in LLMs in real-time.

Findings

Improves biography generation quality by up to 2.5x

Reduces compute consumption by over 6 times

Effective detection and correction of hallucinations during generation

Abstract

In this paper, we present HalluCana, a canary lookahead to detect and correct factuality hallucinations of Large Language Models (LLMs) in long-form generation. HalluCana detects and intervenes as soon as traces of hallucination emerge, during and even before generation. To support timely detection, we exploit the internal factuality representation in the LLM hidden space, where we investigate various proxies to the LLMs' factuality self-assessment, and discuss its relation to the models' context familiarity from their pre-training. On biography generation, our method improves generation quality by up to 2.5x, while consuming over 6 times less compute.