Update README.md
Browse files
README.md
CHANGED
|
@@ -29,12 +29,6 @@ This model is intended for research in:
|
|
| 29 |
- Robust reasoning under adversarial settings
|
| 30 |
- Chain-of-thought alignment studies
|
| 31 |
|
| 32 |
-
## Evaluation
|
| 33 |
-
|
| 34 |
-
The model has been evaluated on:
|
| 35 |
-
- **Safety benchmarks**: StrongReject, BeaverTails
|
| 36 |
-
- **Reasoning benchmarks**: MATH500, GPQA, AIME24
|
| 37 |
-
|
| 38 |
For details, see our [paper](https://arxiv.org/pdf/2505.14667).
|
| 39 |
|
| 40 |
## Overview Results
|
|
|
|
| 29 |
- Robust reasoning under adversarial settings
|
| 30 |
- Chain-of-thought alignment studies
|
| 31 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
For details, see our [paper](https://arxiv.org/pdf/2505.14667).
|
| 33 |
|
| 34 |
## Overview Results
|