File size: 2,268 Bytes
fdf4b05 2e8462d fdf4b05 2e8462d fdf4b05 7d94f5f fdf4b05 7d94f5f fdf4b05 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 |
# Muk-Jji-Bba Dataset (SquidGame Series)
**Note**: Please do not use this dataset for training purposes.
## Overview
The "Muk-Jji-Bba" dataset is the first in the SquidGame series, designed to evaluate whether models can understand human behavior. This dataset specifically focuses on the game of Muk-Jji-Bba, a variation of Rock-Paper-Scissors widely played in Korea.
### How Muk-Jji-Bba Works:
- The attacker tries to match their gesture with the defender’s to win the game.
- If both players show the same gesture in the next round, the attacker wins.
- If the attacker’s next gesture loses to the defender’s, the roles switch, and the defender becomes the new attacker.
- If the attacker’s next gesture beats the defender’s, the attacker keeps their role and the game continues.
The dataset includes 4 rounds per situation. If there is no winner in the final round, the result is a "Tie." The model must choose between Player A (1), Player B (2), or Tie (3).
The labels are evenly distributed across the three possible outcomes.
<style>
table {
width: 60%; /* 표 너비를 60%로 설정 */
font-size: 18px; /* 표의 글자 크기를 12px로 설정 */
margin-left: 0;
margin-right: auto;
}
th, td {
padding: 8px; /* 셀 내부 패딩 설정 */
text-align: center;
}
</style>
## Model Performance
<table>
<tr>
<th>Model</th>
<th>Acc</th>
<th>F1</th>
</tr>
<tr>
<td>Llama3.1-8b</td>
<td>0.808</td>
<td>0.812</td>
</tr>
<tr>
<td>Llama3-8b</td>
<td>0.775</td>
<td>0.778</td>
</tr>
<tr>
<td>Solar-10.7b</td>
<td>0.754</td>
<td>0.759</td>
</tr>
<tr>
<td>Qwen-8b</td>
<td>0.783</td>
<td>0.788</td>
</tr>
<tr>
<td><strong>Yi-chat-9b (TOP)</strong></td>
<td>0.840</td>
<td>0.843</td>
</tr>
</table>
Models with fewer than 12 billion parameters were used due to GPU limitations. 😂
## About the Labels
The labels in the dataset represent the correct outcome for each round:
- Player A wins (1)
- Player B wins (2)
- Tie (3)
The labels are evenly distributed among the three outcomes to ensure balance.
## Stay Tuned
Look forward to the next series in the SquidGame dataset!
(Evaluation code will be uploaded soon.)
|