Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2408.06904 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908412895297536 |
|---|---|
| author | Wang, Zhihu Zhao, Shiwan Wang, Yu Huang, Heyuan Xie, Sitao Zhang, Yubo Shi, Jiaxin Wang, Zhixing Li, Hongyan Yan, Junchi |
| author_facet | Wang, Zhihu Zhao, Shiwan Wang, Yu Huang, Heyuan Xie, Sitao Zhang, Yubo Shi, Jiaxin Wang, Zhixing Li, Hongyan Yan, Junchi |
| contents | The Chain-of-Thought (CoT) paradigm has become a pivotal method for solving complex problems with large language models (LLMs). However, its application to domain-specific tasks remains challenging, as LLMs often fail to decompose tasks accurately or execute subtasks effectively. This paper introduces the Re-TASK framework, a novel theoretical model that revisits LLM tasks from capability, skill, and knowledge perspectives, drawing on the principles of Bloom's Taxonomy and Knowledge Space Theory. While CoT provides a workflow-centric perspective on tasks, Re-TASK introduces a Chain-of-Learning (CoL) paradigm that highlights task dependencies on specific capability items, further broken down into their constituent knowledge and skill components. To address CoT failures, we propose a Re-TASK prompting strategy, which strengthens task-relevant capabilities through targeted knowledge injection and skill adaptation. Experiments across diverse domains demonstrate the effectiveness of Re-TASK. In particular, we achieve improvements of 45.00% on Yi-1.5-9B and 24.50% on Llama3-Chinese-8B for legal tasks. These results highlight the potential of Re-TASK to significantly enhance LLM performance and its applicability in specialized domains. We release our code and data at https://github.com/Uylee/Re-TASK. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_06904 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Re-TASK: Revisiting LLM Tasks from Capability, Skill, and Knowledge Perspectives Wang, Zhihu Zhao, Shiwan Wang, Yu Huang, Heyuan Xie, Sitao Zhang, Yubo Shi, Jiaxin Wang, Zhixing Li, Hongyan Yan, Junchi Computation and Language The Chain-of-Thought (CoT) paradigm has become a pivotal method for solving complex problems with large language models (LLMs). However, its application to domain-specific tasks remains challenging, as LLMs often fail to decompose tasks accurately or execute subtasks effectively. This paper introduces the Re-TASK framework, a novel theoretical model that revisits LLM tasks from capability, skill, and knowledge perspectives, drawing on the principles of Bloom's Taxonomy and Knowledge Space Theory. While CoT provides a workflow-centric perspective on tasks, Re-TASK introduces a Chain-of-Learning (CoL) paradigm that highlights task dependencies on specific capability items, further broken down into their constituent knowledge and skill components. To address CoT failures, we propose a Re-TASK prompting strategy, which strengthens task-relevant capabilities through targeted knowledge injection and skill adaptation. Experiments across diverse domains demonstrate the effectiveness of Re-TASK. In particular, we achieve improvements of 45.00% on Yi-1.5-9B and 24.50% on Llama3-Chinese-8B for legal tasks. These results highlight the potential of Re-TASK to significantly enhance LLM performance and its applicability in specialized domains. We release our code and data at https://github.com/Uylee/Re-TASK. |
| title | Re-TASK: Revisiting LLM Tasks from Capability, Skill, and Knowledge Perspectives |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2408.06904 |