Understanding Generative AI Content with Embedding Models
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_ | 1866912241127784448 |
|---|---|
| author | Vargas, Max Cannon, Reilly Engel, Andrew Sarwate, Anand D. Chiang, Tony |
| author_facet | Vargas, Max Cannon, Reilly Engel, Andrew Sarwate, Anand D. Chiang, Tony |
| contents | Constructing high-quality features is critical to any quantitative data analysis. While feature engineering was historically addressed by carefully hand-crafting data representations based on domain expertise, deep neural networks (DNNs) now offer a radically different approach. DNNs implicitly engineer features by transforming their input data into hidden feature vectors called embeddings. For embedding vectors produced by foundation models -- which are trained to be useful across many contexts -- we demonstrate that simple and well-studied dimensionality-reduction techniques such as Principal Component Analysis uncover inherent heterogeneity in input data concordant with human-understandable explanations. Of the many applications for this framework, we find empirical evidence that there is intrinsic separability between real samples and those generated by artificial intelligence (AI). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_10437 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Understanding Generative AI Content with Embedding Models Vargas, Max Cannon, Reilly Engel, Andrew Sarwate, Anand D. Chiang, Tony Machine Learning Artificial Intelligence Constructing high-quality features is critical to any quantitative data analysis. While feature engineering was historically addressed by carefully hand-crafting data representations based on domain expertise, deep neural networks (DNNs) now offer a radically different approach. DNNs implicitly engineer features by transforming their input data into hidden feature vectors called embeddings. For embedding vectors produced by foundation models -- which are trained to be useful across many contexts -- we demonstrate that simple and well-studied dimensionality-reduction techniques such as Principal Component Analysis uncover inherent heterogeneity in input data concordant with human-understandable explanations. Of the many applications for this framework, we find empirical evidence that there is intrinsic separability between real samples and those generated by artificial intelligence (AI). |
| title | Understanding Generative AI Content with Embedding Models |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2408.10437 |