Name
Abstract | ||
|---|---|---|
Existing approaches for the Table-to-Text task suffer from issues such as missing information, hallucination and repetition. Many approaches to this problem use Reinforcement Learning (RL), which maximizes a single manually defined reward, such as BLEU. In this work, we instead pose the Table-to-Text task as Inverse Reinforcement Learning (IRL) problem. We explore using multiple interpretable unsupervised reward components that are combined linearly to form a composite reward function. The composite reward function and the description generator are learned jointly. We find that IRL outperforms strong RL baselines marginally. We further study the generalization of learned IRL rewards in scenarios involving domain adaptation. Our experiments reveal significant challenges in using IRL for this task. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.18653/v1/2021.acl-short.11 | ACL-IJCNLP 2021: THE 59TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS AND THE 11TH INTERNATIONAL JOINT CONFERENCE ON NATURAL LANGUAGE PROCESSING, VOL 2 |
DocType | Volume | Citations |
Conference | 2021.acl-short | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Sayan Ghosh | 1 | 17 | 8.98 |
| Zheng Qi | 2 | 0 | 0.34 |
| Snigdha Chaturvedi | 3 | 83 | 12.24 |
| Shashank Srivastava | 4 | 0 | 3.04 |