Update README.md

README.md

@@ -17,172 +17,36 @@ tags:
 
 ---
 
+| Benchmark | Ling-mini-2.0 | LLaDA-MoE-7B-A1B-Instruct | LLaDA2.0-mini-preview |
+| :------------------------------ | :-------------: | :-------------------------: | :---------------------: |
+| **Average** | 60.67 | 52.39 | 58.71 |
+| **Knowledge** | | | |
+| MMLU | 78.75 | 67.18 | 72.49 |
+| MMLU-PRO | 56.40 | 44.64 | 49.22 |
+| GPQA | 37.99 | 31.09 | 31.82 |
+| CMMLU | 77.84 | 64.30 | 67.53 |
+| C-EVAL | 77.85 | 63.93 | 66.54 |
+| **Reasoning** | | | |
+| squad2.0 | 69.14 | 86.81 | 85.61 |
+| drop | 76.35 | 79.77 | 79.49 |
+| korbench | 51.04 | 38.40 | 37.26 |
+| **Coding** | | | |
+| CruxEval-O | 71.12 | 42.38 | 61.88 |
+| mbpp | 81.03 | 70.02 | 77.75 |
+| MultiPL-E | 62.23 | 52.53 | 62.43 |
+| humaneval | 77.44 | 61.59 | 80.49 |
+| livecodebench_v6 | 30.18 | 13.27 | 19.93 |
+| Bigcodebench-Full | 35.88 | 20.44 | 30.44 |
+| **Math** | | | |
+| GSM8K | 91.58 | 82.41 | 89.01 |
+| math | 82.22 | 58.68 | 73.50 |
+| OlympiadBench | 49.93 | 21.04 | 36.67 |
+| **Agent & Alignment** | | | |
+| BFCL_Live | 45.74 | 63.09 | 74.11 |
+| IFEval-strict -prompt | 69.13 | 59.33 | 62.50 |
+| SyllogEval<sup>*</sup> | 33.28 | 64.22 | 47.34 |
+| IXRB<sup>*</sup> | 19.00 | 15.00 | 27.00 |
 
-<style type="text/css">
-.tg  {border-collapse:collapse;border-spacing:0;margin:0px auto;}
-.tg td{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;
-  overflow:hidden;padding:10px 5px;word-break:normal;}
-.tg th{border-color:black;border-style:solid;border-width:1px;font-family:Arial, sans-serif;font-size:14px;
-  font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;}
-.tg .tg-baqh{text-align:center;vertical-align:top}
-.tg .tg-0lax{text-align:left;vertical-align:top}
-.tg .tg-amwm{font-weight:bold;text-align:center;vertical-align:top}
-</style>
-<table class="tg"><thead>
-  <tr>
-    <th class="tg-baqh" rowspan="2">AVG</th>
-    <th class="tg-0lax">Ling-mini-2.0</th>
-    <th class="tg-0lax">LLaDA-MoE-7B-A1B-Instruct</th>
-    <th class="tg-0lax">LLaDA2.0-mini-preview</th>
-  </tr>
-  <tr>
-    <th class="tg-amwm">60.67</th>
-    <th class="tg-amwm">52.39</th>
-    <th class="tg-amwm">58.71</th>
-  </tr></thead>
-<tbody>
-  <tr>
-    <td class="tg-amwm" colspan="4">Knowledge</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">MMLU</td>
-    <td class="tg-baqh">78.75</td>
-    <td class="tg-baqh">67.18</td>
-    <td class="tg-baqh">72.49</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">MMLU-PRO</td>
-    <td class="tg-baqh">56.40</td>
-    <td class="tg-baqh">44.64</td>
-    <td class="tg-baqh">49.22</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">GPQA</td>
-    <td class="tg-baqh">37.99</td>
-    <td class="tg-baqh">31.09</td>
-    <td class="tg-baqh">31.82</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">CMMLU</td>
-    <td class="tg-baqh">77.84</td>
-    <td class="tg-baqh">64.30</td>
-    <td class="tg-baqh">67.53</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">C-EVAL</td>
-    <td class="tg-baqh">77.85</td>
-    <td class="tg-baqh">63.93</td>
-    <td class="tg-baqh">66.54</td>
-  </tr>
-  <tr>
-    <td class="tg-amwm" colspan="4">Reasoning</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">squad2.0</td>
-    <td class="tg-baqh">69.14</td>
-    <td class="tg-baqh">86.81</td>
-    <td class="tg-baqh">85.61</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">drop</td>
-    <td class="tg-baqh">76.35</td>
-    <td class="tg-baqh">79.77</td>
-    <td class="tg-baqh">79.49</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">korbench</td>
-    <td class="tg-baqh">51.04</td>
-    <td class="tg-baqh">38.40</td>
-    <td class="tg-baqh">37.26</td>
-  </tr>
-  <tr>
-    <td class="tg-amwm" colspan="4">Coding</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">CruxEval-O</td>
-    <td class="tg-baqh">71.12</td>
-    <td class="tg-baqh">42.38</td>
-    <td class="tg-baqh">61.88</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">mbpp</td>
-    <td class="tg-baqh">81.03</td>
-    <td class="tg-baqh">70.02</td>
-    <td class="tg-baqh">77.75</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">MultiPL-E</td>
-    <td class="tg-baqh">62.23</td>
-    <td class="tg-baqh">52.53</td>
-    <td class="tg-baqh">62.43</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">humaneval</td>
-    <td class="tg-baqh">77.44</td>
-    <td class="tg-baqh">61.59</td>
-    <td class="tg-baqh">80.49</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">livecodebench_v6</td>
-    <td class="tg-baqh">30.18</td>
-    <td class="tg-baqh">13.27</td>
-    <td class="tg-baqh">19.93</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">Bigcodebench-Full</td>
-    <td class="tg-baqh">35.88</td>
-    <td class="tg-baqh">20.44</td>
-    <td class="tg-baqh">30.44</td>
-  </tr>
-  <tr>
-    <td class="tg-amwm" colspan="4">Math</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">GSM8K</td>
-    <td class="tg-baqh">91.58</td>
-    <td class="tg-baqh">82.41</td>
-    <td class="tg-baqh">89.01</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">math</td>
-    <td class="tg-baqh">82.22</td>
-    <td class="tg-baqh">58.68</td>
-    <td class="tg-baqh">73.50</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">OlympiadBench</td>
-    <td class="tg-baqh">49.93</td>
-    <td class="tg-baqh">21.04</td>
-    <td class="tg-baqh">36.67</td>
-  </tr>
-  <tr>
-    <td class="tg-amwm" colspan="4">Agent &amp; Alignment</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">BFCL_Live</td>
-    <td class="tg-baqh">45.74</td>
-    <td class="tg-baqh">63.09</td>
-    <td class="tg-baqh">74.11</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">IFEval-strict -prompt</td>
-    <td class="tg-baqh">69.13</td>
-    <td class="tg-baqh">59.33</td>
-    <td class="tg-baqh">62.50</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">SyllogEval<sup>*</sup></td>
-    <td class="tg-baqh">33.28</td>
-    <td class="tg-baqh">64.22</td>
-    <td class="tg-baqh">47.34</td>
-  </tr>
-  <tr>
-    <td class="tg-0lax">IXRB<sup>*</sup></td>
-    <td class="tg-baqh">19.00</td>
-    <td class="tg-baqh">15.00</td>
-    <td class="tg-baqh">27.00</td>
-  </tr>
-</tbody></table>
 
 **SyllogEval**<sup>*</sup> is a logic benchmark designed to evaluate the formal reasoning capabilities of Large Language Models (LLMs).