Update README.md
Browse files
README.md
CHANGED
|
@@ -53,14 +53,64 @@ We conduct evaluation on both mathematical and coding benchmarks. Due to the hig
|
|
| 53 |
|
| 54 |
</div>
|
| 55 |
|
| 56 |
-
|
| 57 |
-
|
| 58 |
-
|
| 59 |
-
|
| 60 |
-
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
|
| 65 |
|
| 66 |
## Technical Report
|
|
|
|
| 53 |
|
| 54 |
</div>
|
| 55 |
|
| 56 |
+
<table>
|
| 57 |
+
<thead>
|
| 58 |
+
<tr>
|
| 59 |
+
<th rowspan="2">Method</th>
|
| 60 |
+
<th colspan="2">LCB v5 (2024.08.01–2025.02.01)</th>
|
| 61 |
+
<th colspan="2">LCB v6 (2025.02.01–2025.05.01)</th>
|
| 62 |
+
<th rowspan="2">Avg.</th>
|
| 63 |
+
</tr>
|
| 64 |
+
<tr>
|
| 65 |
+
<th>avg@8</th>
|
| 66 |
+
<th>pass@8</th>
|
| 67 |
+
<th>avg@16</th>
|
| 68 |
+
<th>pass@16</th>
|
| 69 |
+
</tr>
|
| 70 |
+
</thead>
|
| 71 |
+
<tbody>
|
| 72 |
+
<tr>
|
| 73 |
+
<td>DeepSeek-R1-1.5B</td>
|
| 74 |
+
<td>16.7</td>
|
| 75 |
+
<td>29.0</td>
|
| 76 |
+
<td>17.2</td>
|
| 77 |
+
<td>34.4</td>
|
| 78 |
+
<td>17.0</td>
|
| 79 |
+
</tr>
|
| 80 |
+
<tr>
|
| 81 |
+
<td>DAPO</td>
|
| 82 |
+
<td>26.0</td>
|
| 83 |
+
<td>40.5</td>
|
| 84 |
+
<td>27.6</td>
|
| 85 |
+
<td>43.5</td>
|
| 86 |
+
<td>26.8</td>
|
| 87 |
+
</tr>
|
| 88 |
+
<tr>
|
| 89 |
+
<td>DeepCoder-1.5B</td>
|
| 90 |
+
<td>23.3</td>
|
| 91 |
+
<td>39.1</td>
|
| 92 |
+
<td>22.6</td>
|
| 93 |
+
<td>42.0</td>
|
| 94 |
+
<td>23.0</td>
|
| 95 |
+
</tr>
|
| 96 |
+
<tr>
|
| 97 |
+
<td>Nemotron-1.5B</td>
|
| 98 |
+
<td>26.1</td>
|
| 99 |
+
<td>35.5</td>
|
| 100 |
+
<td>29.5</td>
|
| 101 |
+
<td>42.8</td>
|
| 102 |
+
<td>27.8</td>
|
| 103 |
+
</tr>
|
| 104 |
+
<tr>
|
| 105 |
+
<td><strong>Archer-Code-1.5B</strong></td>
|
| 106 |
+
<td><strong>29.4</strong></td>
|
| 107 |
+
<td><strong>43.7</strong></td>
|
| 108 |
+
<td><strong>30.2</strong></td>
|
| 109 |
+
<td><strong>45.8</strong></td>
|
| 110 |
+
<td><strong>29.8</strong></td>
|
| 111 |
+
</tr>
|
| 112 |
+
</tbody>
|
| 113 |
+
</table>
|
| 114 |
|
| 115 |
|
| 116 |
## Technical Report
|