Steering Large Language Models to Evaluate and Amplify Creativity
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_ | 1866913602207744000 |
|---|---|
| author | Olson, Matthew Lyle Ratzlaff, Neale Hinck, Musashi Tseng, Shao-yen Lal, Vasudev |
| author_facet | Olson, Matthew Lyle Ratzlaff, Neale Hinck, Musashi Tseng, Shao-yen Lal, Vasudev |
| contents | Although capable of generating creative text, Large Language Models (LLMs) are poor judges of what constitutes "creativity". In this work, we show that we can leverage this knowledge of how to write creatively in order to better judge what is creative. We take a mechanistic approach that extracts differences in the internal states of an LLM when prompted to respond "boringly" or "creatively" to provide a robust measure of creativity that corresponds strongly with human judgment. We also show these internal state differences can be applied to enhance the creativity of generated text at inference time. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_06060 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Steering Large Language Models to Evaluate and Amplify Creativity Olson, Matthew Lyle Ratzlaff, Neale Hinck, Musashi Tseng, Shao-yen Lal, Vasudev Computation and Language Artificial Intelligence Although capable of generating creative text, Large Language Models (LLMs) are poor judges of what constitutes "creativity". In this work, we show that we can leverage this knowledge of how to write creatively in order to better judge what is creative. We take a mechanistic approach that extracts differences in the internal states of an LLM when prompted to respond "boringly" or "creatively" to provide a robust measure of creativity that corresponds strongly with human judgment. We also show these internal state differences can be applied to enhance the creativity of generated text at inference time. |
| title | Steering Large Language Models to Evaluate and Amplify Creativity |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2412.06060 |