Helping Language Models Learn More: Multi-dimensional Task Prompt for Few-shot Tuning
This work addresses the challenge of effectively utilizing specific knowledge in LLMs like ChatGPT for few-shot tuning, though it appears incremental as it builds on existing prompt learning principles.
The paper tackles the problem of prompt uncertainty in large language models by proposing MTPrompt, a multi-dimensional task prompt learning method that automatically builds and searches for appropriate prompts, achieving the best results on few-shot samples across five datasets.
Large language models (LLMs) can be used as accessible and intelligent chatbots by constructing natural language queries and directly inputting the prompt into the large language model. However, different prompt' constructions often lead to uncertainty in the answers and thus make it hard to utilize the specific knowledge of LLMs (like ChatGPT). To alleviate this, we use an interpretable structure to explain the prompt learning principle in LLMs, which certificates that the effectiveness of language models is determined by position changes of the task's related tokens. Therefore, we propose MTPrompt, a multi-dimensional task prompt learning method consisting based on task-related object, summary, and task description information. By automatically building and searching for appropriate prompts, our proposed MTPrompt achieves the best results on few-shot samples setting and five different datasets. In addition, we demonstrate the effectiveness and stability of our method in different experimental settings and ablation experiments. In interaction with large language models, embedding more task-related information into prompts will make it easier to stimulate knowledge embedded in large language models.