File size: 7,004 Bytes
a2c11fe
 
62dfebc
2dd6ecd
ceb9fbf
a2c11fe
8936a9e
 
 
a2c11fe
62dfebc
a2c11fe
55d723f
609959b
2c87331
 
 
ceb9fbf
d6029bb
4bf0072
a412524
 
3b63af9
a412524
 
 
 
 
 
 
ceb9fbf
55d723f
a2c11fe
ceb9fbf
a2c11fe
 
 
 
18d1dac
ceb9fbf
 
62dfebc
452352b
ceb9fbf
62dfebc
ceb9fbf
 
62dfebc
 
ceb9fbf
 
a2c11fe
 
 
 
cebd12d
fac91f3
a2c11fe
ceb9fbf
a2c11fe
 
 
ceb9fbf
a2c11fe
 
8936a9e
62dfebc
 
ceb9fbf
 
8936a9e
b442b7b
 
 
ceb9fbf
62dfebc
 
 
 
 
 
 
 
 
 
 
a2c11fe
 
 
 
 
d6029bb
ceb9fbf
d6029bb
8731413
ceb9fbf
 
 
62dfebc
 
 
 
 
 
 
 
 
 
 
 
 
ceb9fbf
 
 
62dfebc
ceb9fbf
 
62dfebc
ceb9fbf
 
 
d6029bb
8731413
409694a
2bcdab6
 
8731413
1b4bbb2
 
8731413
 
 
17e3ef5
818db54
8731413
d6029bb
8731413
ceb9fbf
d6029bb
 
 
818db54
1b4bbb2
 
 
7396151
1b4bbb2
7396151
1b4bbb2
007bbde
a2c11fe
 
62dfebc
ceb9fbf
 
62dfebc
 
 
 
ceb9fbf
a2c11fe
62dfebc
 
 
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
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
# Dataset Card for JamC-QA

[英語](https://huggingface.co/datasets/sbintuitions/JamC-QA)/日本語

## Dataset Summary

このベンチマークは、日本の文化や風習といった日本固有の知識を問う多肢選択式の質問応答ベンチマークです。
文化(culture)、風習(custom)、風土(regional_identity)、地理(geography)、日本史(history)、行政(government)、法律(law)、医療(healthcare)の 8つのカテゴリの問題を含みます。
このベンチマークで高いスコアを達成するには、日本に関する幅広い知識を獲得している必要があります。

## Leaderboard

### Evaluation Metric

本評価では選択肢に付与されたラベル(0-3)でなく、LLM に選択肢文字列そのものを出力させています。
以下のスコアは LLM の出力した文字列が正解選択肢の文字列と完全一致した問題数の割合です。


| モデル | All | 文化 | 風習 | 風土 | 地理 | 日本史 | 行政 | 法律 | 医療 | 
|:---|----|---:|---:|---:|---:|---:|---:|---:|---:|
| [sarashina2-8x70b](https://huggingface.co/sbintuitions/sarashina2-8x70b) | **0.725** | 0.714 | **0.775** | **0.761** | 0.654 | **0.784** | 0.736 | 0.632 | **0.917** |
| [sarashina2-70b](https://huggingface.co/sbintuitions/sarashina2-70b) | **0.725** | **0.719** | 0.745 | 0.736 | **0.673** | 0.764 | 0.764 | 0.666 | **0.917** |
| [Llama-3.3-Swallow-70B-v0.4](https://huggingface.co/tokyotech-llm/Llama-3.3-Swallow-70B-v0.4) | 0.697 | 0.689 | **0.775** | 0.589 | 0.566 | 0.776 | **0.773** | **0.783** | 0.854 |
| [RakutenAI-2.0-8x7B](https://huggingface.co/Rakuten/RakutenAI-2.0-8x7B) | 0.633 | 0.622 | 0.725 | 0.617 | 0.511 | 0.714 | 0.709 | 0.575 | 0.813 |
| [plamo-100b](https://huggingface.co/pfnet/plamo-100b) | 0.603 | 0.602 | 0.650 | 0.637 | 0.504 | 0.682 | 0.609 | 0.515 | 0.688 |
| [Mixtral-8x7B-v0.1-japanese](https://huggingface.co/abeja/Mixtral-8x7B-v0.1-japanese) | 0.593 | 0.602 | 0.670 | 0.579 | 0.493 | 0.612 | 0.736 | 0.545 | 0.667 |
| [Meta-Llama-3.1-405B](https://huggingface.co/meta-llama/Llama-3.1-405B) | 0.571 | 0.558 | 0.545 | 0.484 | 0.500 | 0.679 | 0.646 | 0.629 | 0.688 |
| [llm-jp-3.1-8x13b](https://huggingface.co/llm-jp/llm-jp-3-8x13b) | 0.568 | 0.595 | 0.635 | 0.582 | 0.449 | 0.589 | 0.627 | 0.502 | 0.625 |
| [Nemotron-4-340B-Base](https://huggingface.co/mgoin/Nemotron-4-340B-Base-hf) | 0.567 | 0.573 | 0.615 | 0.511 | 0.467 | 0.595 | 0.727 | 0.582 | 0.667 |
| [Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B) | 0.527 | 0.522 | 0.595 | 0.426 | 0.438 | 0.606 | 0.609 | 0.562 | 0.688 |

## Language

日本語

## Dataset Structure
### Data Instances

文化カテゴリの問題例は以下の通り:
```
{
  "qid": "jamcqa-test-culture-00001",
  "category": "culture",
  "question": "「狂った世で気が狂うなら気は確かだ」の名言を残した映画はどれ?",
  "choice0": "影武者",
  "choice1": "羅生門",
  "choice2": "隠し砦の三悪人",
  "choice3": "乱",
  "answer_index": 3,
}
```

## Data Fields

- `qid (str)`: 質問を一意識別するための ID
- `category (str)`: 質問カテゴリ 全8種
  - 文化(culture)、風習(custom)、風土(regional_identity)、地理(geography)、日本史(history)、行政(government)、法律(law)、医療(healthcare)
- `question (str)`: 質問文
  - カタカナを除き、全角→半角統一済み
  - 内部に改行 `\n` を含まない
  - 文字列前後の空白は除去済み
- `choice{0..3} (str)`: 選択肢(choice0〜choice3 の 4つ)
  - カタカナを除き、全角→半角統一済み
  - 内部に改行 `\n` を含まない
  - 文字列前後の空白は除去済み
- `answer_index (int)`: `choice0` 〜`choice3` に対応した正解選択肢のインデックス(0-3)


## Data Splits

- `dev`: カテゴリごとに 4-shot 学習用の 4問を設けています。
- `test`: 全 2,309問。

内訳は以下の通り: 

| カテゴリ | dev | test | 
| --- | ---: | ---: |
| 文化(culture) | 4 | 640 | 
| 風習(custom) | 4 | 200 |
| 風土(regional_identity) | 4 | 397 |
| 地理(geography) | 4 | 272 |
| 日本史(history) | 4 | 343 |
| 行政(government) | 4 | 110 |
| 法律(law) | 4 | 299 |
| 医療(healthcare) | 4 | 48 |
| 合計 | 32 | 2,309 |

# Licensing Information

- [CC-BY-SA-4.0](https://creativecommons.org/licenses/by-sa/4.0/)

# Usage

## Dataset Loading

```python
$ python
>>> import datasets
>>> jamcqa = datasets.load_dataset('sbintuitions/JamC-QA', 'v1.0')
>>> print(jamcqa)
DatasetDict({
    test: Dataset({
        features: ['qid', 'category', 'question', 'choice0', 'choice1', 'choice2', 'choice3', 'answer_index'],
        num_rows: 2309
    })
    dev: Dataset({
        features: ['qid', 'category', 'question', 'choice0', 'choice1', 'choice2', 'choice3', 'answer_index'],
        num_rows: 32
    })
})
>>> jamcqa_test = jamcqa['test']
>>> print(jamcqa_test)
Dataset({
    features: ['qid', 'category', 'question', 'choice0', 'choice1', 'choice2', 'choice3', 'answer_index'],
    num_rows: 2309
})
>>> print(jamcqa_test[0])
{'qid': 'jamcqa-test-culture-00001', 'category': 'culture', 'question': '「狂った世で気が狂うなら気は確かだ」の名言を残した映画はどれ?', 'choice0': '影武者', 'choice1': '羅生門', 'choice2': '隠し砦の三悪人', 'choice3': '乱', 'answer_index': 3}
>>> 
```

## Evaluation with FlexEval

評価用ツール [FlexEval](https://github.com/sbintuitions/flexeval) (>=0.13.3) を使うと、
[こちらのクイックスタート](https://github.com/sbintuitions/flexeval/blob/main/README_ja.md#%E3%82%AF%E3%82%A4%E3%83%83%E3%82%AF%E3%82%B9%E3%82%BF%E3%83%BC%E3%83%88)の
`aio``jamcqa` に置き換えるだけで簡単に JamC-QA のスコア評価を実行できます。

### Run Command

```python
flexeval_lm \
  --language_model HuggingFaceLM \
  --language_model.model "sbintuitions/sarashina2.2-0.5b" \
  --language_model.default_gen_kwargs "{ do_sample: false }" \
  --eval_setup "jamcqa" \
  --save_dir "results/jamcqa"
```

`--language_model.default_gen_kwargs "{ do_sample: false }"` はサンプリングを使わず
[greedy search](https://huggingface.co/docs/transformers/ja/generation_strategies#greedy-search)
を行うオプションです。

### Output

```
...
2025-09-03 15:48:24.633 | INFO     | flexeval.core.evaluate_generation:evaluate_generation:92 - {'exact_match': 0.2368990905153746, 'finish_reason_ratio-stop': 1.0, 'avg_output_length': 6.94283239497618, 'max_output_length': 93, 'min_output_length': 2}
...
```

# Citation Information
```
@inproceedings{Oka2025,
  author={岡 照晃, 柴田 知秀, 吉田 奈央},
  title={JamC-QA: 日本固有の知識を問う多肢選択式質問応答ベンチマークの構築},
  year={2025},
  month={March},
  booktitle={言語処理学会第31回年次大会(NLP2025)},
  pages={839--844},
}
```