CANDLE: Iterative Conceptualization and Instantiation Distillation from Large Language Models for Commonsense Reasoning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914804302610432 |
|---|---|
| author | Wang, Weiqi Fang, Tianqing Li, Chunyang Shi, Haochen Ding, Wenxuan Xu, Baixuan Wang, Zhaowei Bai, Jiaxin Liu, Xin Cheng, Jiayang Chan, Chunkit Song, Yangqiu |
| author_facet | Wang, Weiqi Fang, Tianqing Li, Chunyang Shi, Haochen Ding, Wenxuan Xu, Baixuan Wang, Zhaowei Bai, Jiaxin Liu, Xin Cheng, Jiayang Chan, Chunkit Song, Yangqiu |
| contents | The sequential process of conceptualization and instantiation is essential to generalizable commonsense reasoning as it allows the application of existing knowledge to unfamiliar scenarios. However, existing works tend to undervalue the step of instantiation and heavily rely on pre-built concept taxonomies and human annotations to collect both types of knowledge, resulting in a lack of instantiated knowledge to complete reasoning, high cost, and limited scalability. To tackle these challenges, we introduce CANDLE, a distillation framework that iteratively performs contextualized conceptualization and instantiation over commonsense knowledge bases by instructing large language models to generate both types of knowledge with critic filtering. By applying CANDLE to ATOMIC, we construct a comprehensive knowledge base comprising six million conceptualizations and instantiated commonsense knowledge triples. Both types of knowledge are firmly rooted in the original ATOMIC dataset, and intrinsic evaluations demonstrate their exceptional quality and diversity. Empirical results indicate that distilling CANDLE on student models provides benefits across four downstream tasks. Our code, data, and models are publicly available at https://github.com/HKUST-KnowComp/CANDLE. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_07286 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CANDLE: Iterative Conceptualization and Instantiation Distillation from Large Language Models for Commonsense Reasoning Wang, Weiqi Fang, Tianqing Li, Chunyang Shi, Haochen Ding, Wenxuan Xu, Baixuan Wang, Zhaowei Bai, Jiaxin Liu, Xin Cheng, Jiayang Chan, Chunkit Song, Yangqiu Computation and Language The sequential process of conceptualization and instantiation is essential to generalizable commonsense reasoning as it allows the application of existing knowledge to unfamiliar scenarios. However, existing works tend to undervalue the step of instantiation and heavily rely on pre-built concept taxonomies and human annotations to collect both types of knowledge, resulting in a lack of instantiated knowledge to complete reasoning, high cost, and limited scalability. To tackle these challenges, we introduce CANDLE, a distillation framework that iteratively performs contextualized conceptualization and instantiation over commonsense knowledge bases by instructing large language models to generate both types of knowledge with critic filtering. By applying CANDLE to ATOMIC, we construct a comprehensive knowledge base comprising six million conceptualizations and instantiated commonsense knowledge triples. Both types of knowledge are firmly rooted in the original ATOMIC dataset, and intrinsic evaluations demonstrate their exceptional quality and diversity. Empirical results indicate that distilling CANDLE on student models provides benefits across four downstream tasks. Our code, data, and models are publicly available at https://github.com/HKUST-KnowComp/CANDLE. |
| title | CANDLE: Iterative Conceptualization and Instantiation Distillation from Large Language Models for Commonsense Reasoning |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2401.07286 |