Update README.md
Browse files
README.md
CHANGED
@@ -65,6 +65,12 @@ Simplify your answer as much as possible.
|
|
65 |
{Instruction} [/INST]
|
66 |
```
|
67 |
|
|
|
|
|
|
|
|
|
|
|
|
|
68 |
## Citation
|
69 |
```
|
70 |
@misc{yuan2024advancing,
|
|
|
65 |
{Instruction} [/INST]
|
66 |
```
|
67 |
|
68 |
+
## Evaluation
|
69 |
+
- Eurus, both the 7B and 70B variants, achieve the best overall performance among open-source models of similar sizes. Eurus even outperforms specialized models in corresponding domains in many cases. Notably, Eurus-7B outperforms baselines that are 5× larger, and Eurus-70B achieves better performance than GPT-3.5 Turbo.
|
70 |
+
- Preference learning with UltraInteract can further improve performance, especially in math and the multi-turn ability.
|
71 |
+
<img src="./figures/main_exp.png" alt="stats" style="zoom: 40%;" />
|
72 |
+
|
73 |
+
|
74 |
## Citation
|
75 |
```
|
76 |
@misc{yuan2024advancing,
|