REVIEW 3 cited by
Natural Language Generation for Advertising: A Survey
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Natural language generation methods have emerged as effective tools to help advertisers increase the number of online advertisements they produce. This survey entails a review of the research trends on this topic over the past decade, from template-based to extractive and abstractive approaches using neural networks. Additionally, key challenges and directions revealed through the survey, including metric optimization, faithfulness, diversity, multimodality, and the development of benchmark datasets, are discussed.
Forward citations
Cited by 3 Pith papers
-
CTR-Driven Ad Text Generation via Online Feedback Preference Optimization
CTOP combines retrieval-augmented style transfer with DPO weighted by CTR gain and AA-group confidence, achieving +4.76% relative CTR over human-written ad titles in online tests.
-
AdParaphrase v2.0: Generating Attractive Ad Texts Using a Preference-Annotated Paraphrase Dataset
AdParaphrase v2.0 provides 16,460 preference-annotated Japanese ad paraphrase pairs, revealing linguistic correlates of attractiveness and supporting DPO-based rewriting, though labels are noisy and online evidence is...
-
Exploring the Relationship Between Diversity and Quality in Ad Text Generation
In Japanese ad text generation, methods that increase output diversity generally lower ad quality, with beam search and sampling responding differently to examples and output counts.
Discussion (0). Sign in to comment.