This repository contains data for our paper "AdParaphrase: Paraphrase Dataset for Analyzing Linguistic Features toward Generating Attractive Ad Texts" (NAACL2025 Findings).

Overview

AdParaphrase is a novel paraphrase dataset that contains human preferences for pairs of ad texts that are semantically equivalent but differ in wording and style. We carefully constructed the dataset by collecting semantically similar ad texts, performing… See the full description on the dataset page: https://huggingface.co/datasets/cyberagent/AdParaphrase.