Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.08454 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929755300823040 |
|---|---|
| author | Chan, Zhangming Chen, Xiuying Wang, Yongliang Li, Juntao Zhang, Zhiqiang Gai, Kun Zhao, Dongyan Yan, Rui |
| author_facet | Chan, Zhangming Chen, Xiuying Wang, Yongliang Li, Juntao Zhang, Zhiqiang Gai, Kun Zhao, Dongyan Yan, Rui |
| contents | Different from other text generation tasks, in product description generation, it is of vital importance to generate faithful descriptions that stick to the product attribute information. However, little attention has been paid to this problem. To bridge this gap, we propose a model named Fidelity-oriented Product Description Generator (FPDG). FPDG takes the entity label of each word into account, since the product attribute information is always conveyed by entity words. Specifically, we first propose a Recurrent Neural Network (RNN) decoder based on the Entity-label-guided Long Short-Term Memory (ELSTM) cell, taking both the embedding and the entity label of each word as input. Second, we establish a keyword memory that stores the entity labels as keys and keywords as values, allowing FPDG to attend to keywords by attending to their entity labels. Experiments conducted on a large-scale real-world product description dataset show that our model achieves state-of-the-art performance in terms of both traditional generation metrics and human evaluations. Specifically, FPDG increases the fidelity of the generated descriptions by 25%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_08454 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Stick to Facts: Towards Fidelity-oriented Product Description Generation Chan, Zhangming Chen, Xiuying Wang, Yongliang Li, Juntao Zhang, Zhiqiang Gai, Kun Zhao, Dongyan Yan, Rui Computation and Language Different from other text generation tasks, in product description generation, it is of vital importance to generate faithful descriptions that stick to the product attribute information. However, little attention has been paid to this problem. To bridge this gap, we propose a model named Fidelity-oriented Product Description Generator (FPDG). FPDG takes the entity label of each word into account, since the product attribute information is always conveyed by entity words. Specifically, we first propose a Recurrent Neural Network (RNN) decoder based on the Entity-label-guided Long Short-Term Memory (ELSTM) cell, taking both the embedding and the entity label of each word as input. Second, we establish a keyword memory that stores the entity labels as keys and keywords as values, allowing FPDG to attend to keywords by attending to their entity labels. Experiments conducted on a large-scale real-world product description dataset show that our model achieves state-of-the-art performance in terms of both traditional generation metrics and human evaluations. Specifically, FPDG increases the fidelity of the generated descriptions by 25%. |
| title | Stick to Facts: Towards Fidelity-oriented Product Description Generation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2503.08454 |