Uncovering Implicit Bias in Large Language Models with Concept Learning Dataset
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909925208227840 |
|---|---|
| author | Wang, Leroy Z. |
| author_facet | Wang, Leroy Z. |
| contents | We introduce a dataset of concept learning tasks that helps uncover implicit biases in large language models. Using in-context concept learning experiments, we found that language models may have a bias toward upward monotonicity in quantifiers; such bias is less apparent when the model is tested by direct prompting without concept learning components. This demonstrates that in-context concept learning can be an effective way to discover hidden biases in language models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_01219 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Uncovering Implicit Bias in Large Language Models with Concept Learning Dataset Wang, Leroy Z. Computation and Language Artificial Intelligence We introduce a dataset of concept learning tasks that helps uncover implicit biases in large language models. Using in-context concept learning experiments, we found that language models may have a bias toward upward monotonicity in quantifiers; such bias is less apparent when the model is tested by direct prompting without concept learning components. This demonstrates that in-context concept learning can be an effective way to discover hidden biases in language models. |
| title | Uncovering Implicit Bias in Large Language Models with Concept Learning Dataset |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2510.01219 |