Spaces:
No application file
No application file
Update README.md
Browse files
README.md
CHANGED
|
@@ -12,7 +12,7 @@ colorTo: green
|
|
| 12 |
sdk_version: 6.9.0
|
| 13 |
---
|
| 14 |
|
| 15 |
-
#
|
| 16 |
|
| 17 |
**Active UltraFeedback** is a scalable pipeline for generating high-quality preference datasets to align large language models (LLMs). We leverage **uncertainty quantification** and **active learning** to annotate only the most informative samples, drastically reducing costs while beating standard baselines.
|
| 18 |
|
|
|
|
| 12 |
sdk_version: 6.9.0
|
| 13 |
---
|
| 14 |
|
| 15 |
+
# This repo accompanies the paper: [ActiveUltraFeedback — arXiv:2603.09692](https://arxiv.org/abs/2603.09692).
|
| 16 |
|
| 17 |
**Active UltraFeedback** is a scalable pipeline for generating high-quality preference datasets to align large language models (LLMs). We leverage **uncertainty quantification** and **active learning** to annotate only the most informative samples, drastically reducing costs while beating standard baselines.
|
| 18 |
|