Context-Tuning: Learning Contextualized Prompts for Natural Language Generation
This addresses the issue of insufficient input semantics consideration in prompting methods for natural language generation, offering a lightweight solution for improved text generation relevance.
The paper tackles the problem of input-independent prompting in pretrained language models for natural language generation by proposing context-tuning, which derives prompts from input text and uses continuous inverse prompting to enhance relevance, achieving performance with only 0.12% of parameters fine-tuned.
Recently, pretrained language models (PLMs) have had exceptional success in language generation. To leverage the rich knowledge encoded by PLMs, a simple yet powerful paradigm is to use prompts in the form of either discrete tokens or continuous embeddings. In existing studies, these prompting methods are typically independent of the inputs, lacking sufficient consideration of input semantics. To address this issue, we propose a novel continuous prompting approach, called context-tuning, to fine-tuning PLMs for natural language generation. Firstly, the prompts are derived based on the input text to elicit useful knowledge from PLMs for generation. We refer to such prompts as contextualized prompts. Secondly, we use continuous inverse prompting to improve the process of natural language generation by modeling an inverse generation process from output to input, making the generated text more relevant to the inputs. Furthermore, we utilize a lightweight context-tuning method that fine-tunes only 0.12% of the parameters while maintaining good performance. Our code is publicly available at https://github.com/RUCAIBox/Context-Tuning.