beyondguo/LLM-Tuning
Academic research introducing Sample Design Engineering (SDE) to systematically improve LLM downstream fine-tuning through better training sample design.
Not currently ranked — collecting fresh signals.
star history
The repository presents empirical research on downstream fine-tuning of large language models, introducing the concept of Sample Design Engineering (SDE). Through comprehensive experiments across multiple LLMs, the work investigates various sample design strategies and their effects on fine-tuning performance. The researchers develop ES-SDE, an approach that integrates effective design options, and demonstrate its superiority over baseline methods on new tasks.
Frequently asked
- What is beyondguo/LLM-Tuning?
- Academic research introducing Sample Design Engineering (SDE) to systematically improve LLM downstream fine-tuning through better training sample design.
- Is LLM-Tuning open source?
- Yes — beyondguo/LLM-Tuning is an open-source project tracked on heatdrop.
- What language is LLM-Tuning written in?
- beyondguo/LLM-Tuning is primarily written in HTML.
- How popular is LLM-Tuning?
- beyondguo/LLM-Tuning has 1k stars on GitHub.
- Where can I find LLM-Tuning?
- beyondguo/LLM-Tuning is on GitHub at https://github.com/beyondguo/LLM-Tuning.