Name
Abstract | ||
|---|---|---|
Generalized text representations are the foundation of many natural language understanding tasks. To fully utilize the different corpus, it is inevitable that models need to understand the relevance among them. However, many methods ignore the relevance and adopt a single-channel model (a coarse paradigm) directly for all tasks, which lacks enough rationality and interpretation. In addition, some existing works learn downstream tasks by stitches skill block (a fine paradigm), which might cause irrational results due to its redundancy and noise. In this work, we first analyze the task correlation through three different perspectives, , data property, manual design, and model-based relevance, based on which the similar tasks are grouped together. Then, we propose a hierarchical framework with a coarse-to-fine paradigm, with the bottom level shared to all the tasks, the mid-level divided to different groups, and the top-level assigned to each of the tasks. This allows our model to learn basic language properties from all tasks, boost performance on relevant tasks, and reduce the negative impact from irrelevant tasks. Our experiments on 13 benchmark datasets across five natural language understanding tasks demonstrate the superiority of our method. |
Year | Venue | DocType |
|---|---|---|
2022 | International Conference on Computational Linguistics | Conference |
Volume | Citations | PageRank |
Proceedings of the 29th International Conference on Computational Linguistics | 0 | 0.34 |
References | Authors | |
0 | 12 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zhaoye Fei | 1 | 0 | 0.34 |
| Yu Tian | 2 | 0 | 3.04 |
| Yongkang Wu | 3 | 0 | 0.34 |
| Xinyu Zhang | 4 | 8 | 3.53 |
| Yutao Zhu | 5 | 0 | 1.01 |
| Liu Zheng | 6 | 47 | 12.80 |
| Jiawen Wu | 7 | 0 | 0.68 |
| Dejiang Kong | 8 | 4 | 1.74 |
| Ruofei Lai | 9 | 0 | 0.34 |
| Zhao Cao | 10 | 6 | 3.85 |
| Zhicheng Dou | 11 | 706 | 41.96 |
| Xipeng Qiu | 12 | 556 | 63.33 |