Update Gen-Eval scores to latest version (#3)
Browse files- Update Gen-Eval scores to latest version (fda0d4b1721f2b5ba6ebbaecdde439ab7b546e77)
Co-authored-by: Audrey <AudreyVM@users.noreply.huggingface.co>
README.md
CHANGED
@@ -767,39 +767,38 @@ provided with MT-GenEval.
|
|
767 |
|
768 |
<details>
|
769 |
|
770 |
-
|
|
771 |
-
|
772 |
-
|
|
773 |
-
| SalamandraTA-7b-base
|
774 |
-
|
|
775 |
-
| TowerInstruct-7B-v0.2
|
776 |
| | | | | | |
|
777 |
-
|
|
778 |
-
| SalamandraTA-7b-base
|
779 |
-
|
|
780 |
-
| TowerInstruct-7B-v0.2
|
781 |
| | | | | | |
|
782 |
-
|
|
783 |
-
| SalamandraTA-7b-base
|
784 |
-
|
|
785 |
-
| TowerInstruct-7B-v0.2
|
786 |
| | | | | | |
|
787 |
-
|
|
788 |
-
| SalamandraTA-7b-base
|
789 |
-
|
|
790 |
-
| TowerInstruct-7B-v0.2
|
791 |
| | | | | | |
|
792 |
-
|
|
793 |
-
| SalamandraTA-7b-base
|
794 |
-
|
|
795 |
-
| TowerInstruct-7B-v0.2
|
|
|
|
|
|
|
|
|
|
|
796 |
| | | | | | |
|
797 |
-
| SalamandraTA-7b-instruct | en | ru | **0.95** | **0.837** | **0.793** |
|
798 |
-
| SalamandraTA-7b-base | en | ru | 0.933 | 0.713 | 0.653 |
|
799 |
-
| MADLAD400-7B-mt | en | ru | 0.94 | 0.797 | 0.74 |
|
800 |
-
| TowerInstruct-7B-v0.2 | en | ru | 0.933 | 0.797 | 0.733 |
|
801 |
-
|
802 |
-
<img src="./images/geneval.png"/>
|
803 |
|
804 |
</details>
|
805 |
|
|
|
767 |
|
768 |
<details>
|
769 |
|
770 |
+
| | Source | Target | Masc | Fem | Pair |
|
771 |
+
|:--|:--|:--|:--|:--|:--|
|
772 |
+
| MADLAD400-7B | en | de | **0.877** | 0.823 | 0.713 |
|
773 |
+
| SalamandraTA-7b-base | en | de | 0.857 | 0.770 | 0.660 |
|
774 |
+
| SalamandraTA-7b-instruct | en | de | 0.863 | **0.867** | **0.740** |
|
775 |
+
| TowerInstruct-7B-v0.2 | en | de | 0.863 | 0.840 | 0.727 |
|
776 |
| | | | | | |
|
777 |
+
| MADLAD400-7B | en | es | 0.887 | 0.780 | 0.687 |
|
778 |
+
| SalamandraTA-7b-base | en | es | **0.890** | 0.733 | 0.643 |
|
779 |
+
| SalamandraTA-7b-instruct | en | es | 0.860 | **0.837** | **0.710** |
|
780 |
+
| TowerInstruct-7B-v0.2 | en | es | 0.850 | 0.823 | 0.693 |
|
781 |
| | | | | | |
|
782 |
+
| MADLAD400-7B | en | fr | 0.873 | 0.777 | 0.663 |
|
783 |
+
| SalamandraTA-7b-base | en | fr | 0.887 | 0.710 | 0.617 |
|
784 |
+
| SalamandraTA-7b-instruct | en | fr | **0.900** | 0.813 | **0.730** |
|
785 |
+
| TowerInstruct-7B-v0.2 | en | fr | 0.880 | **0.823** | 0.717 |
|
786 |
| | | | | | |
|
787 |
+
| MADLAD400-7B | en | it | 0.907 | 0.663 | 0.597 |
|
788 |
+
| SalamandraTA-7b-base | en | it | 0.893 | 0.593 | 0.513 |
|
789 |
+
| SalamandraTA-7b-instruct | en | it | 0.913 | **0.780** | 0.707 |
|
790 |
+
| TowerInstruct-7B-v0.2 | en | it | **0.947** | 0.747 | **0.713** |
|
791 |
| | | | | | |
|
792 |
+
| MADLAD400-7B | en | pt | 0.923 | 0.687 | 0.627 |
|
793 |
+
| SalamandraTA-7b-base | en | pt | 0.923 | 0.650 | 0.597 |
|
794 |
+
| SalamandraTA-7b-instruct | en | pt | **0.933** | **0.797** | **0.747** |
|
795 |
+
| TowerInstruct-7B-v0.2 | en | pt | 0.907 | 0.730 | 0.670 |
|
796 |
+
| | | | | | |
|
797 |
+
| MADLAD400-7B | en | ru | 0.940 | 0.797 | 0.740 |
|
798 |
+
| SalamandraTA-7b-base | en | ru | 0.933 | 0.713 | 0.653 |
|
799 |
+
| SalamandraTA-7b-instruct | en | ru | **0.950** | **0.830** | **0.783** |
|
800 |
+
| TowerInstruct-7B-v0.2 | en | ru | 0.933 | 0.797 | 0.733 |
|
801 |
| | | | | | |
|
|
|
|
|
|
|
|
|
|
|
|
|
802 |
|
803 |
</details>
|
804 |
|