SCOPE: A Self-supervised Framework for Improving Faithfulness in Conditional Text Generation

TL;DR

This paper introduces SCOPE, a self-supervised framework that enhances faithfulness in conditional text generation by training models to prefer grounded outputs, significantly reducing hallucinations in tasks like summarization.

Contribution

SCOPE presents a novel self-supervised training method that generates unfaithful samples to improve faithfulness in conditional text generation models.

Findings

Outperforms existing self-supervised methods in faithfulness metrics

Reduces hallucinations in summarization and data-to-text tasks

Improves human and automatic evaluation scores for groundedness

Abstract

Large Language Models (LLMs), when used for conditional text generation, often produce hallucinations, i.e., information that is unfaithful or not grounded in the input context. This issue arises in typical conditional text generation tasks, such as text summarization and data-to-text generation, where the goal is to produce fluent text based on contextual input. When fine-tuned on specific domains, LLMs struggle to provide faithful answers to a given context, often adding information or generating errors. One underlying cause of this issue is that LLMs rely on statistical patterns learned from their training data. This reliance can interfere with the model's ability to stay faithful to a provided context, leading to the generation of ungrounded information. We build upon this observation and introduce a novel self-supervised method for generating a training set of unfaithful samples. We then refine the model using a training process that encourages the generation of grounded outputs over unfaithful ones, drawing on preference-based training. Our approach leads to significantly more grounded text generation, outperforming existing self-supervised techniques in faithfulness, as evaluated through automatic metrics, LLM-based assessments, and human evaluations.