Papers › Copy mechanism and tailored training for character-based data-to-text generation

Copy mechanism and tailored training for character-based data-to-text generation

26 Apr 2019arXiv:1904.11838archive 2025-07-28

Marco Roberti, Giovanni Bonetta, Rossella Cancelliere, Patrick Gallinari

In the last few years, many different methods have been focusing on using deep recurrent neural networks for natural language generation. The most widely used sequence-to-sequence neural methods are word-based: as such, they need a pre-processing step called delexicalization (conversely, relexicalization) to deal with uncommon or unknown words. These forms of processing, however, give rise to models that depend on the vocabulary used and are not completely neural. In this work, we present an end-to-end sequence-to-sequence model with attention mechanism which reads and generates at a character level, no longer requiring delexicalization, tokenization, nor even lowercasing. Moreover, since characters constitute the common "building blocks" of every text, it also allows a more general approach to text generation, enabling the possibility to exploit transfer learning for training. These skills are obtained thanks to two major features: (i) the possibility to alternate between the standard generation mechanism and a copy one, which allows to directly copy input facts to produce outputs, and (ii) the use of an original training pipeline that further improves the quality of the generated texts. We also introduce a new dataset called E2E+, designed to highlight the copying capabilities of character-based models, that is a modified version of the well-known E2E dataset used in the E2E Challenge. We tested our model according to five broadly accepted metrics (including the widely used BLEU), showing that it yields competitive performance with respect to both character-based and word-based approaches.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

marco-roberti/char-data-to-text-gen officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

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

Tasks

Data-to-Text GenerationText GenerationTransfer Learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Data-to-Text Generation E2E NLG Challenge EDA_CS BLEU 67.05 #2 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS CIDEr 2.2355 #2 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS METEOR 44.49 #2 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS NIST 8.5150 #2 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS ROUGE-L 68.94 #2 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS (TL) BLEU 65.80 #6 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS (TL) CIDEr 2.1803 #6 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS (TL) METEOR 45.16 #6 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS (TL) NIST 8.5615 #6 of 11 Archive leaderboard report
Data-to-Text Generation E2E NLG Challenge EDA_CS (TL) ROUGE-L 67.40 #6 of 11 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