A library-event bulletin summarizer wins a reference-overlap contest by copying sentences, while readers prefer a shorter rewrite that a human panel ranks higher. What do those two scores mean?
Select an answer to reveal the explanation.
Short Explanation
ROUGE can crown a copy-heavy draft. Readers can still prefer a shorter rewrite. Overlap measures n-gram match to a gold summary; a human panel measures usefulness. Those two scoreboards can disagree, and the library’s question is the reader panel.
Full Explanation
ROUGE-style n-gram overlap asks how much a candidate shares with a reference summary. A copy-heavy draft can win that contest and still be a poor bulletin if readers wanted a shorter rewrite. Human preference answers usefulness, and the two scoreboards can disagree. The library's question is the reader panel, reported beside, not instead of, the overlap number.