Papers › May the Force Be with Your Copy Mechanism: Enhanced Supervised-Copy Method for Natural...

May the Force Be with Your Copy Mechanism: Enhanced Supervised-Copy Method for Natural Language Generation

20 Dec 2021arXiv 2021 12arXiv:2112.10360archive 2025-07-28

Sanghyuk Choi, Jeong-in Hwang, Hyungjong Noh, Yeonsoo Lee

Recent neural sequence-to-sequence models with a copy mechanism have achieved remarkable progress in various text generation tasks. These models addressed out-of-vocabulary problems and facilitated the generation of rare words. However, the identification of the word which needs to be copied is difficult, as observed by prior copy models, which suffer from incorrect generation and lacking abstractness. In this paper, we propose a novel supervised approach of a copy network that helps the model decide which words need to be copied and which need to be generated. Specifically, we re-define the objective function, which leverages source sequences and target vocabularies as guidance for copying. The experimental results on data-to-text generation and abstractive summarization tasks verify that our approach enhances the copying quality and improves the degree of abstractness.

PaperPDFConference PDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Abstractive Text SummarizationData-to-Text GenerationText Generation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Data-to-Text Generation MLB Dataset Force-Copy BLEU 10.5 #4 of 4 Archive leaderboard report
Data-to-Text Generation MLB Dataset (Content Ordering) Force-Copy DLD 21.16 #3 of 4 Archive leaderboard report
Data-to-Text Generation MLB Dataset (Content Selection) Force-Copy Precision 49.39 #1 of 3 Archive leaderboard report
Data-to-Text Generation MLB Dataset (Content Selection) Force-Copy Recall 50.89 #1 of 3 Archive leaderboard report
Data-to-Text Generation MLB Dataset (Relation Generation) Force-Copy Precision 84.50 #3 of 4 Archive leaderboard report
Data-to-Text Generation MLB Dataset (Relation Generation) Force-Copy count 21.05 #3 of 4 Archive leaderboard report
Data-to-Text Generation RotoWire Force-Copy BLEU 17.26 #3 of 6 Archive leaderboard report
Data-to-Text Generation RotoWire (Content Ordering) Force-Copy BLEU 15.8 #4 of 5 Archive leaderboard report
Data-to-Text Generation RotoWire (Content Ordering) Force-Copy DLD 17.26% #4 of 5 Archive leaderboard report
Data-to-Text Generation RotoWire (Relation Generation) Force-Copy Precision 95.40% #3 of 6 Archive leaderboard report
Data-to-Text Generation RotoWire (Relation Generation) Force-Copy count 27.37 #3 of 6 Archive leaderboard report
Data-to-Text Generation Rotowire (Content Selection) Force-Copy Precision 34.34% #2 of 5 Archive leaderboard report
Data-to-Text Generation Rotowire (Content Selection) Force-Copy Recall 48.85% #2 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections