Datasets:

jaagli
/

othello_shuffle

step stringclasses 61 values	num_black int64 0 64	num_white int64 0 64	game_id stringlengths 5 7	curr_player stringclasses 3 values
55	26	33	126687	black
18	9	13	126687	white
54	24	34	126687	white
32	17	19	126687	white
3	5	2	126687	black
35	21	18	126687	black
51	29	26	126687	black
27	20	11	126687	black
4	4	4	126687	white
7	8	3	126687	black
44	23	25	126687	white
12	5	11	126687	white
52	22	34	126687	white
56	20	40	126687	white
0	2	2	126687	begin
22	11	15	126687	white
46	23	27	126687	white
2	3	3	126687	white
1	4	1	126687	black
17	10	11	126687	black
50	25	29	126687	white
47	25	26	126687	black
11	6	9	126687	black
20	11	13	126687	white
9	6	7	126687	black
23	14	13	126687	black
14	6	12	126687	white
41	30	15	126687	black
40	23	21	126687	white
31	20	15	126687	black
38	20	22	126687	white
42	27	19	126687	white
16	7	13	126687	white
26	15	15	126687	white
58	21	41	126687	white
30	18	16	126687	white
19	12	11	126687	black
29	21	12	126687	black
21	16	9	126687	black
5	6	3	126687	black
13	7	10	126687	black
15	9	10	126687	black
39	25	18	126687	black
33	20	17	126687	black
24	13	15	126687	white
34	17	21	126687	white
49	26	27	126687	black
36	19	21	126687	white
25	16	13	126687	black
45	29	20	126687	black
53	30	27	126687	black
10	4	10	126687	white
6	5	5	126687	white
48	20	32	126687	white
57	23	38	126687	black
60	22	42	126687	white
37	21	20	126687	black
43	31	16	126687	black
28	19	13	126687	white
59	23	40	126687	black
8	4	8	126687	white
55	41	18	775719	black
18	5	17	775719	white
54	35	23	775719	white
32	20	16	775719	white
3	5	2	775719	black
35	26	13	775719	black
51	38	17	775719	black
27	18	13	775719	black
4	3	5	775719	white
7	6	5	775719	black
44	28	20	775719	white
12	5	11	775719	white
52	37	19	775719	white
56	39	21	775719	white
0	2	2	775719	begin
22	11	15	775719	white
46	31	19	775719	white
2	3	3	775719	white
1	4	1	775719	black
17	8	13	775719	black
50	33	21	775719	white
47	36	15	775719	black
11	6	9	775719	black
20	6	18	775719	white
9	7	6	775719	black
23	13	14	775719	black
14	4	14	775719	white
41	34	11	775719	black
40	28	16	775719	white
31	21	14	775719	black
38	27	15	775719	white
42	31	15	775719	white
16	6	14	775719	white
26	13	17	775719	white
58	43	19	775719	white
30	17	17	775719	white
19	7	16	775719	black
29	22	11	775719	black
21	12	13	775719	black

End of preview. Expand in Data Studio

Dataset Card

Dataset Description

The dataset is derived from real Othello game records collected from EOTHELLO. It combines textual move sequences with corresponding visual board states, enabling joint modeling of language and vision in a structured, rule-based environment.

Each game consists of a sequence of 60 ± 2 moves on average, with one board image generated after every move. This results in a total of approximately 25,000 games and 1.56 million board images.

It provides two synchronized modalities:

Text modality – move tokens representing board positions (e.g., “C4”, “E6”).
Visual modality – RGB images depicting the full Othello board state after each move.

Repository:

multimodal-othello

Statistics

Split	Number of Games	Number of Images	Avg. Images per Game
Training	20,525	1,247,852	~ 60.8
Validation	1,282	78,141	~ 60.9
Test	3,850	233,975	~ 60.8
Total	25,657	1,559,968	~ 60.8

Intended Usage

The dataset is intended for academic research for training and/or evaluating language models.

Citation

Paper: What if Othello-Playing Language Models Could See?

BibTeX:

@article{chen2025if,
  title={What if Othello-Playing Language Models Could See?},
  author={Chen, Xinyi and Yuan, Yifei and Li, Jiaang and Belongie, Serge and de Rijke, Maarten and S{\o}gaard, Anders},
  journal={arXiv preprint arXiv:2507.14520},
  year={2025}
}

Downloads last month: 67

Size of downloaded dataset files:

34.8 GB

Size of the auto-converted Parquet files:

34.8 GB

Number of rows:

1,559,968