URL: Universal Referential Knowledge Linking via Task-instructed Representation Compression
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913328922624000 |
|---|---|
| author | Li, Zhuoqun Lin, Hongyu Wang, Tianshu Cao, Boxi Lu, Yaojie Zhou, Weixiang Wang, Hao Zeng, Zhenyu Sun, Le Han, Xianpei |
| author_facet | Li, Zhuoqun Lin, Hongyu Wang, Tianshu Cao, Boxi Lu, Yaojie Zhou, Weixiang Wang, Hao Zeng, Zhenyu Sun, Le Han, Xianpei |
| contents | Linking a claim to grounded references is a critical ability to fulfill human demands for authentic and reliable information. Current studies are limited to specific tasks like information retrieval or semantic matching, where the claim-reference relationships are unique and fixed, while the referential knowledge linking (RKL) in real-world can be much more diverse and complex. In this paper, we propose universal referential knowledge linking (URL), which aims to resolve diversified referential knowledge linking tasks by one unified model. To this end, we propose a LLM-driven task-instructed representation compression, as well as a multi-view learning approach, in order to effectively adapt the instruction following and semantic understanding abilities of LLMs to referential knowledge linking. Furthermore, we also construct a new benchmark to evaluate ability of models on referential knowledge linking tasks across different scenarios. Experiments demonstrate that universal RKL is challenging for existing approaches, while the proposed framework can effectively resolve the task across various scenarios, and therefore outperforms previous approaches by a large margin. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_16248 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | URL: Universal Referential Knowledge Linking via Task-instructed Representation Compression Li, Zhuoqun Lin, Hongyu Wang, Tianshu Cao, Boxi Lu, Yaojie Zhou, Weixiang Wang, Hao Zeng, Zhenyu Sun, Le Han, Xianpei Computation and Language Artificial Intelligence Linking a claim to grounded references is a critical ability to fulfill human demands for authentic and reliable information. Current studies are limited to specific tasks like information retrieval or semantic matching, where the claim-reference relationships are unique and fixed, while the referential knowledge linking (RKL) in real-world can be much more diverse and complex. In this paper, we propose universal referential knowledge linking (URL), which aims to resolve diversified referential knowledge linking tasks by one unified model. To this end, we propose a LLM-driven task-instructed representation compression, as well as a multi-view learning approach, in order to effectively adapt the instruction following and semantic understanding abilities of LLMs to referential knowledge linking. Furthermore, we also construct a new benchmark to evaluate ability of models on referential knowledge linking tasks across different scenarios. Experiments demonstrate that universal RKL is challenging for existing approaches, while the proposed framework can effectively resolve the task across various scenarios, and therefore outperforms previous approaches by a large margin. |
| title | URL: Universal Referential Knowledge Linking via Task-instructed Representation Compression |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2404.16248 |