Attribute First, then Generate: Locally-attributable Grounded Text Generation

arXiv:2403.17104v1 Announce Type: new
Abstract: Recent efforts to address hallucinations in Large Language Models (LLMs) have focused on attributed text generation, which supplements generated texts with citations of supporting sources for post-generation fact-checking and corrections. Yet, these citations often point to entire documents or paragraphs, burdening users with extensive verification work. In this paper, we introduce a locally-attributable text generation approach, prioritizing concise attributions. Our method, named ``Attribute First, then Generate'', breaks down the conventional end-to-end generation process into three …

abstract arxiv citations cs.cl documents fact-checking generate generated hallucinations language language models large language large language models llms text text generation type verification work