Refine Knowledge of Large Language Models via Adaptive Contrastive Learning
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916608186777600 |
|---|---|
| author | Li, Yinghui Huang, Haojing Kuang, Jiayi Li, Yangning Guo, Shu-Yu Qu, Chao Tan, Xiaoyu Zheng, Hai-Tao Shen, Ying Yu, Philip S. |
| author_facet | Li, Yinghui Huang, Haojing Kuang, Jiayi Li, Yangning Guo, Shu-Yu Qu, Chao Tan, Xiaoyu Zheng, Hai-Tao Shen, Ying Yu, Philip S. |
| contents | How to alleviate the hallucinations of Large Language Models (LLMs) has always been the fundamental goal pursued by the LLMs research community. Looking through numerous hallucination-related studies, a mainstream category of methods is to reduce hallucinations by optimizing the knowledge representation of LLMs to change their output. Considering that the core focus of these works is the knowledge acquired by models, and knowledge has long been a central theme in human societal progress, we believe that the process of models refining knowledge can greatly benefit from the way humans learn. In our work, by imitating the human learning process, we design an Adaptive Contrastive Learning strategy. Our method flexibly constructs different positive and negative samples for contrastive learning based on LLMs' actual mastery of knowledge. This strategy helps LLMs consolidate the correct knowledge they already possess, deepen their understanding of the correct knowledge they have encountered but not fully grasped, forget the incorrect knowledge they previously learned, and honestly acknowledge the knowledge they lack. Extensive experiments and detailed analyses on widely used datasets demonstrate the effectiveness of our method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_07184 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Refine Knowledge of Large Language Models via Adaptive Contrastive Learning Li, Yinghui Huang, Haojing Kuang, Jiayi Li, Yangning Guo, Shu-Yu Qu, Chao Tan, Xiaoyu Zheng, Hai-Tao Shen, Ying Yu, Philip S. Computation and Language Artificial Intelligence How to alleviate the hallucinations of Large Language Models (LLMs) has always been the fundamental goal pursued by the LLMs research community. Looking through numerous hallucination-related studies, a mainstream category of methods is to reduce hallucinations by optimizing the knowledge representation of LLMs to change their output. Considering that the core focus of these works is the knowledge acquired by models, and knowledge has long been a central theme in human societal progress, we believe that the process of models refining knowledge can greatly benefit from the way humans learn. In our work, by imitating the human learning process, we design an Adaptive Contrastive Learning strategy. Our method flexibly constructs different positive and negative samples for contrastive learning based on LLMs' actual mastery of knowledge. This strategy helps LLMs consolidate the correct knowledge they already possess, deepen their understanding of the correct knowledge they have encountered but not fully grasped, forget the incorrect knowledge they previously learned, and honestly acknowledge the knowledge they lack. Extensive experiments and detailed analyses on widely used datasets demonstrate the effectiveness of our method. |
| title | Refine Knowledge of Large Language Models via Adaptive Contrastive Learning |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2502.07184 |