Browse State-of-the-Art › Data-to-Text Generation

Data-to-Text Generation

112 papers with code · 26 benchmarks · 24 datasets archive 2025-07-28

Natural Language Processing

A classic problem in natural-language generation (NLG) involves taking structured data, such as a table, as input, and producing text that adequately and fluently describes this data as output. Unlike machine translation, which aims for complete transduction of the sentence to be translated, this form of NLG is usually taken to require addressing (at least) two separate challenges: what to say, the selection of an appropriate subset of the input data to discuss, and how to say it, the surface realization of a generation.

( Image credit: Data-to-Text Generation with Content Selection and Planning )

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

26 leaderboard tables shown for this task, 26 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 26 until expanded.

Syntology 1 newer paper was judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): RotoWire (1 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
WebNLG (20 rows) Control Prefixes (A1, T5-large) Control Prefixes for Parameter-Efficient Text Generation code — Compare
E2E NLG Challenge (11 rows) S_1^R Pragmatically Informative Text Generation code — Compare
WebNLG Full (8 rows) Control Prefixes (A1, A2, T5-large) Control Prefixes for Parameter-Efficient Text Generation code — Compare
Cleaned E2E NLG Challenge (7 rows) Control Prefixes (T5-large) Control Prefixes for Parameter-Efficient Text Generation code — Compare
RotoWire (6 rows) HierarchicalEncoder + NR + IR Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text... code — Compare
RotoWire (Relation Generation) (6 rows) SeqPlan Data-to-text Generation with Variational Sequential Planning code — Compare
ToTTo (6 rows) T5-3B Text-to-Text Pre-Training for Data-to-Text Tasks code — Compare
XAlign (6 rows) Fact-aware embedding with mT5 XF2T: Cross-lingual Fact-to-Text Generation for Low-Resource Languages — — Compare
DART (5 rows) T5B Baseline FactSpotter: Evaluating the Factual Faithfulness of Graph-to-Text... code — Compare
MULTIWOZ 2.1 (5 rows) T5-Base Text-to-Text Pre-Training for Data-to-Text Tasks code — Compare
RotoWire (Content Ordering) (5 rows) Hierarchical Transformer Encoder + conditional copy A Hierarchical Model for Data-to-Text Generation code — Compare
Rotowire (Content Selection) (5 rows) Hierarchical Transformer Encoder + conditional copy A Hierarchical Model for Data-to-Text Generation code — Compare
MLB Dataset (Relation Generation) (4 rows) SeqPlan Data-to-text Generation with Variational Sequential Planning code — Compare
MLB Dataset (4 rows) SeqPlan Data-to-text Generation with Variational Sequential Planning code — Compare
MLB Dataset (Content Ordering) (4 rows) SeqPlan Data-to-text Generation with Variational Sequential Planning code — Compare
Czech Restaurant NLG (3 rows) binmt Machine Translation Pre-training for Data-to-Text Generation -- A... — — Compare
MLB Dataset (Content Selection) (3 rows) Force-Copy May the Force Be with Your Copy Mechanism: Enhanced... — — Compare
E2E (2 rows) self-mem + new data (random) Self-training from Self-memory in Data-to-text Generation code — Compare
SR11Deep (2 rows) Transition based Deep Input Linearization Transition-Based Deep Input Linearization code — Compare
ViGGO (2 rows) DataTuner_FC Have Your Text and Use It Too! End-to-End Neural Data-to-Text... code — Compare
WebNLG en (2 rows) mBART The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics — — Compare
WebNLG ru (2 rows) mBART The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics — — Compare
AMR3.0 (1 row) StructAdapt Structural Adapters in Pretrained Language Models for AMR-to-text... code — Compare
GenWiki (1 row) T5-large Ontology-Free General-Domain Knowledge Graph-to-Text Generation... code Syntology official (archive's flag): 5 ran · 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified Compare
WikiOFGraph (1 row) T5-large Ontology-Free General-Domain Knowledge Graph-to-Text Generation... code Syntology official (archive's flag): 5 ran · 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified Compare
Wikipedia Person and Animal Dataset (1 row) Ours Towards Faithful Neural Table-to-Text Generation with... — — Compare

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

24 datasets whose archive record lists this task, ordered by the archive's paper count.

Subtasks archive 2025-07-28

3 subtasks in the archive's task tree.

Parent tasks archive 2025-07-28

Papers archive 2025-07-28

archive papers tagged: 219 · with a code link: 112 · where Syntology ran a sample: 22 (16 with a run with no instrument failure, 6 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (22 of 219 tagged: 16 with a run with no instrument failure, 6 where every run was a failure of Syntology's instrument)

Page 1 of 3: papers 1 to 100 of 219, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 21 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

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