STAR : Sentence Translation Alignment Rate for Document-to-Document Machine Translation
Read the original on arXiv Computation and Language →The paper introduces STAR, a metric that measures sentence-level alignment between source and target documents in document-to-document machine translation. Using STAR, the authors develop StarPO, a preference‑optimization framework that ranks translation hypotheses by structural quality and applies a dynamic alignment mask to focus learning on misaligned segments. Experiments on news and literary data show that StarPO improves both translation quality and structural integrity, enabling small models to outperform large proprietary systems such as GPT‑4o while remaining more token‑efficient.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.