ANAH-v2：擴展大型語言模型的分析性幻覺標註

ANAH-v2: Scaling Analytical Hallucination Annotation of Large Language Models

July 5, 2024

作者: Yuzhe Gu, Ziwei Ji, Wenwei Zhang, Chengqi Lyu, Dahua Lin, Kai Chen

cs.AI

摘要

大型語言模型（LLMs）在各個領域和廣泛應用的長文問答任務中表現出幻覺。目前的幻覺檢測和緩解數據集在領域和大小上受限，由於勞動成本高昂和現有幻覺標註者的可靠性不足，很難擴展。為了促進對LLM幻覺的可擴展監督，本文介紹了一個迭代自我訓練框架，同時逐步擴大幻覺標註數據集的規模，並提高幻覺標註者的準確性。基於期望最大化（EM）算法，在每個迭代中，該框架首先應用一個幻覺標註流程對一個擴大的數據集進行標註，然後在該數據集上訓練一個更準確的幻覺標註者。這個新的幻覺標註者被採用在下一次迭代中使用的幻覺標註流程中。廣泛的實驗結果表明，最終獲得的僅具有7B參數的幻覺標註者超越了GPT-4的性能，在HaluEval和HalluQA上實現了零-shot推理的新的最先進幻覺檢測結果。這樣的標註者不僅可以評估大規模數據集上各種LLMs的幻覺水平，還可以幫助減輕LLMs生成的幻覺，使自然語言推理（NLI）指標從25%提高到37%。

English

Large language models (LLMs) exhibit hallucinations in long-form question-answering tasks across various domains and wide applications. Current hallucination detection and mitigation datasets are limited in domains and sizes, which struggle to scale due to prohibitive labor costs and insufficient reliability of existing hallucination annotators. To facilitate the scalable oversight of LLM hallucinations, this paper introduces an iterative self-training framework that simultaneously and progressively scales up the hallucination annotation dataset and improves the accuracy of the hallucination annotator. Based on the Expectation Maximization (EM) algorithm, in each iteration, the framework first applies a hallucination annotation pipeline to annotate a scaled dataset and then trains a more accurate hallucination annotator on the dataset. This new hallucination annotator is adopted in the hallucination annotation pipeline used for the next iteration. Extensive experimental results demonstrate that the finally obtained hallucination annotator with only 7B parameters surpasses the performance of GPT-4 and obtains new state-of-the-art hallucination detection results on HaluEval and HalluQA by zero-shot inference. Such an annotator can not only evaluate the hallucination levels of various LLMs on the large-scale dataset but also help to mitigate the hallucination of LLMs generations, with the Natural Language Inference (NLI) metric increasing from 25% to 37% on HaluEval.