Round to 2 decimal digits
Browse files
README.md
CHANGED
|
@@ -46,19 +46,19 @@ Below we report recoveries on individual tasks as well as the average recovery.
|
|
| 46 |
|
| 47 |
| Task | Recovery (%) |
|
| 48 |
|------|--------------|
|
| 49 |
-
| SingleOp | 97.
|
| 50 |
-
| SingleQ | 95.
|
| 51 |
-
| MultiArith | 98.
|
| 52 |
-
| SVAMP | 95.
|
| 53 |
-
| GSM8K | 93.
|
| 54 |
-
| MMLU-Math | 80.
|
| 55 |
-
| BBH-LogicalDeduction-3Obj | 89.
|
| 56 |
-
| BBH-ObjectCounting | 82.
|
| 57 |
-
| BBH-Navigate | 90.
|
| 58 |
-
| TabFact | 86.
|
| 59 |
-
| HotpotQA | 96.
|
| 60 |
-
| SQuAD | 98.
|
| 61 |
-
| DROP | 94.
|
| 62 |
-
| Winograd-WSC | 89.
|
| 63 |
-
| Average | **92.
|
| 64 |
|
|
|
|
| 46 |
|
| 47 |
| Task | Recovery (%) |
|
| 48 |
|------|--------------|
|
| 49 |
+
| SingleOp | 97.94 |
|
| 50 |
+
| SingleQ | 95.95 |
|
| 51 |
+
| MultiArith | 98.22 |
|
| 52 |
+
| SVAMP | 95.08 |
|
| 53 |
+
| GSM8K | 93.69 |
|
| 54 |
+
| MMLU-Math | 80.54 |
|
| 55 |
+
| BBH-LogicalDeduction-3Obj | 89.87 |
|
| 56 |
+
| BBH-ObjectCounting | 82.03 |
|
| 57 |
+
| BBH-Navigate | 90.66 |
|
| 58 |
+
| TabFact | 86.92 |
|
| 59 |
+
| HotpotQA | 96.81 |
|
| 60 |
+
| SQuAD | 98.46 |
|
| 61 |
+
| DROP | 94.33 |
|
| 62 |
+
| Winograd-WSC | 89.47 |
|
| 63 |
+
| Average | **92.14** |
|
| 64 |
|