{"paper_id":"33df18a2-3571-49d9-b28a-49fa5b969f01","body_text":"Abstract\nLarge language models excel at text extraction, but they sometimes hallucinate. A simple way to avoid hallucinations is to remove any extracted text that does not appear in the original source. This is easy when the extracted text is contiguous (findable with exact string matching), but much harder when it is discontiguous. Techniques for finding discontiguous phrases depend heavily on how the text is split—i.e., how it is tokenized. In this study, we show that splitting text along subword boundaries, with LLM-specific tokenizers, and aligning extracted text with ordered alignment algorithms, improves alignment by about 50% compared to word-level tokenization. To demonstrate this, we introduce the Berkeley Ordered Alignment of Text (BOAT) dataset, a modification of the Stanford Question Answering Dataset (SQuAD) that includes non-contiguous phrases, and BIO-BOAT a biomedical variant built from 51 bioRxiv preprints. We show that text-alignment methods form a partially ordered set, and that ordered alignment is the most practical choice for verifying LLM-extracted text. We implement this approach in taln, which enumerates ordinal subword alignments.\nCompeting Interest Statement\nThe authors have declared no competing interest.\nFootnotes\nCorrected the manuscript title, added an Author Contribution section, updated the Acknowledgements Section, and made a few minor edits.","source_license":"CC-BY-4.0","license_restricted":false}