Direct-Inverse Prompting: Analyzing LLMs' Discriminative Capacity in Self-Improving Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929421823246336 |
|---|---|
| author | Ahn, Jihyun Janice Kamoi, Ryo Cheng, Lu Zhang, Rui Yin, Wenpeng |
| author_facet | Ahn, Jihyun Janice Kamoi, Ryo Cheng, Lu Zhang, Rui Yin, Wenpeng |
| contents | Mainstream LLM research has primarily focused on enhancing their generative capabilities. However, even the most advanced LLMs experience uncertainty in their outputs, often producing varied results on different runs or when faced with minor changes in input, despite no substantial change in content. Given multiple responses from the same LLM to the same input, we advocate leveraging the LLMs' discriminative capability to reduce this generative uncertainty, aiding in identifying the correct answers. Specifically, we propose and analyze three discriminative prompts: direct, inverse, and hybrid, to explore the potential of both closed-source and open-source LLMs in self-improving their generative performance on two benchmark datasets. Our insights reveal which discriminative prompt is most promising and when to use it. To our knowledge, this is the first work to systematically analyze LLMs' discriminative capacity to address generative uncertainty. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_11017 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Direct-Inverse Prompting: Analyzing LLMs' Discriminative Capacity in Self-Improving Generation Ahn, Jihyun Janice Kamoi, Ryo Cheng, Lu Zhang, Rui Yin, Wenpeng Computation and Language Artificial Intelligence Machine Learning Mainstream LLM research has primarily focused on enhancing their generative capabilities. However, even the most advanced LLMs experience uncertainty in their outputs, often producing varied results on different runs or when faced with minor changes in input, despite no substantial change in content. Given multiple responses from the same LLM to the same input, we advocate leveraging the LLMs' discriminative capability to reduce this generative uncertainty, aiding in identifying the correct answers. Specifically, we propose and analyze three discriminative prompts: direct, inverse, and hybrid, to explore the potential of both closed-source and open-source LLMs in self-improving their generative performance on two benchmark datasets. Our insights reveal which discriminative prompt is most promising and when to use it. To our knowledge, this is the first work to systematically analyze LLMs' discriminative capacity to address generative uncertainty. |
| title | Direct-Inverse Prompting: Analyzing LLMs' Discriminative Capacity in Self-Improving Generation |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2407.11017 |