Dataset Viewer
Auto-converted to Parquet Duplicate
benchmark_format_version
stringclasses
1 value
dataset_id
stringclasses
1 value
split
stringclasses
1 value
sample_id
stringlengths
9
9
seed
int64
0
499
clues
int64
30
30
size
int64
9
9
box_rows
int64
3
3
box_cols
int64
3
3
puzzle
listlengths
9
9
solution
listlengths
9
9
canonical_steps
listlengths
51
51
difficulty_signature
unknown
sudoku-verify-benchmark-v1
v1
test
test_0000
0
30
9
3
3
[ [ 0, 0, 0, 0, 9, 3, 1, 0, 0 ], [ 0, 0, 0, 7, 0, 0, 0, 0, 5 ], [ 1, 5, 9, 2, 6, 0, 0, 0, 0 ], [ 2, 0, 0, 0, 0, 5, 6, 0, 0 ], [ 5, 0, 4, 0, ...
[ [ 8, 7, 2, 5, 9, 3, 1, 4, 6 ], [ 3, 4, 6, 7, 1, 8, 2, 9, 5 ], [ 1, 5, 9, 2, 6, 4, 8, 7, 3 ], [ 2, 9, 7, 1, 3, 5, 6, 8, 4 ], [ 5, 1, 4, 6, ...
[ { "technique": "naked_single", "row": 5, "col": 2, "value": 3, "candidates": [ 3 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 3, "col": 2, "value": 7, "candidates...
{ "hidden_single": 11, "naked_single": 40 }
sudoku-verify-benchmark-v1
v1
test
test_0001
1
30
9
3
3
[ [ 6, 3, 0, 2, 0, 0, 4, 0, 7 ], [ 0, 0, 0, 1, 0, 6, 3, 0, 2 ], [ 0, 0, 1, 0, 0, 0, 0, 0, 5 ], [ 0, 0, 0, 0, 7, 2, 0, 9, 0 ], [ 0, 1, 0, 0, ...
[ [ 6, 3, 9, 2, 5, 8, 4, 1, 7 ], [ 4, 5, 7, 1, 9, 6, 3, 8, 2 ], [ 8, 2, 1, 7, 3, 4, 9, 6, 5 ], [ 3, 6, 4, 5, 7, 2, 8, 9, 1 ], [ 7, 1, 8, 4, ...
[ { "technique": "naked_single", "row": 1, "col": 7, "value": 8, "candidates": [ 8 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 0, "col": 7, "value": 1, "candidates...
{ "naked_single": 51 }
sudoku-verify-benchmark-v1
v1
test
test_0002
2
30
9
3
3
[ [ 0, 9, 0, 0, 2, 0, 0, 6, 0 ], [ 0, 2, 8, 0, 0, 5, 0, 0, 0 ], [ 0, 0, 6, 1, 0, 4, 8, 0, 7 ], [ 8, 0, 7, 0, 0, 6, 4, 9, 0 ], [ 0, 0, 0, 7, ...
[ [ 4, 9, 1, 8, 2, 7, 5, 6, 3 ], [ 7, 2, 8, 3, 6, 5, 1, 4, 9 ], [ 3, 5, 6, 1, 9, 4, 8, 2, 7 ], [ 8, 1, 7, 5, 3, 6, 4, 9, 2 ], [ 2, 4, 5, 7, ...
[ { "technique": "naked_single", "row": 2, "col": 0, "value": 3, "candidates": [ 3 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 2, "col": 1, "value": 5, "candidates...
{ "hidden_single": 2, "naked_single": 49 }
sudoku-verify-benchmark-v1
v1
test
test_0003
3
30
9
3
3
[ [ 0, 1, 4, 0, 0, 0, 0, 0, 5 ], [ 6, 0, 0, 0, 0, 0, 0, 2, 0 ], [ 0, 5, 0, 0, 0, 2, 6, 0, 0 ], [ 0, 0, 2, 6, 0, 7, 5, 0, 0 ], [ 0, 9, 0, 0, ...
[ [ 2, 1, 4, 8, 7, 6, 9, 3, 5 ], [ 6, 7, 8, 9, 3, 5, 1, 2, 4 ], [ 9, 5, 3, 4, 1, 2, 6, 8, 7 ], [ 3, 8, 2, 6, 9, 7, 5, 4, 1 ], [ 1, 9, 5, 2, ...
[ { "technique": "naked_single", "row": 0, "col": 6, "value": 9, "candidates": [ 9 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 0, "col": 3, "value": 8, "candidates...
{ "naked_single": 51 }
sudoku-verify-benchmark-v1
v1
test
test_0004
4
30
9
3
3
[ [ 0, 9, 3, 0, 0, 0, 0, 4, 5 ], [ 0, 0, 6, 9, 2, 0, 0, 0, 0 ], [ 1, 8, 0, 0, 7, 0, 9, 0, 0 ], [ 0, 3, 0, 2, 0, 0, 0, 0, 8 ], [ 0, 4, 2, 0, ...
[ [ 2, 9, 3, 6, 8, 1, 7, 4, 5 ], [ 5, 7, 6, 9, 2, 4, 3, 8, 1 ], [ 1, 8, 4, 5, 7, 3, 9, 2, 6 ], [ 7, 3, 5, 2, 9, 6, 4, 1, 8 ], [ 8, 4, 2, 1, ...
[ { "technique": "naked_single", "row": 1, "col": 1, "value": 7, "candidates": [ 7 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 0, "col": 0, "value": 2, "candidates...
{ "naked_single": 51 }
sudoku-verify-benchmark-v1
v1
test
test_0005
5
30
9
3
3
[ [ 0, 6, 0, 8, 0, 0, 2, 0, 1 ], [ 0, 8, 4, 0, 3, 2, 0, 0, 6 ], [ 0, 0, 9, 0, 0, 0, 8, 0, 3 ], [ 5, 0, 0, 0, 4, 3, 0, 0, 0 ], [ 0, 0, 0, 0, ...
[ [ 3, 6, 5, 8, 9, 7, 2, 4, 1 ], [ 7, 8, 4, 1, 3, 2, 5, 9, 6 ], [ 1, 2, 9, 4, 5, 6, 8, 7, 3 ], [ 5, 7, 1, 9, 4, 3, 6, 8, 2 ], [ 8, 4, 6, 5, ...
[ { "technique": "naked_single", "row": 0, "col": 5, "value": 7, "candidates": [ 7 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 0, "col": 0, "value": 3, "candidates...
{ "naked_single": 51 }
sudoku-verify-benchmark-v1
v1
test
test_0006
6
30
9
3
3
[ [ 6, 0, 8, 0, 9, 2, 0, 0, 0 ], [ 0, 7, 0, 3, 0, 0, 0, 0, 0 ], [ 1, 0, 0, 7, 0, 0, 4, 5, 0 ], [ 5, 9, 0, 0, 0, 0, 8, 7, 0 ], [ 0, 0, 0, 5, ...
[ [ 6, 5, 8, 4, 9, 2, 7, 3, 1 ], [ 9, 7, 4, 3, 1, 5, 2, 6, 8 ], [ 1, 3, 2, 7, 6, 8, 4, 5, 9 ], [ 5, 9, 1, 2, 3, 6, 8, 7, 4 ], [ 4, 2, 7, 5, ...
[ { "technique": "naked_single", "row": 0, "col": 7, "value": 3, "candidates": [ 3 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 0, "col": 8, "value": 1, "candidates...
{ "naked_single": 51 }
sudoku-verify-benchmark-v1
v1
test
test_0007
7
30
9
3
3
[ [ 0, 0, 4, 3, 6, 0, 8, 7, 9 ], [ 3, 0, 1, 9, 0, 0, 0, 0, 0 ], [ 0, 8, 0, 0, 0, 0, 0, 0, 5 ], [ 6, 0, 0, 1, 0, 0, 0, 0, 7 ], [ 8, 0, 0, 0, ...
[ [ 2, 5, 4, 3, 6, 1, 8, 7, 9 ], [ 3, 7, 1, 9, 5, 8, 2, 4, 6 ], [ 9, 8, 6, 7, 2, 4, 1, 3, 5 ], [ 6, 4, 5, 1, 8, 2, 3, 9, 7 ], [ 8, 1, 3, 6, ...
[ { "technique": "naked_single", "row": 0, "col": 1, "value": 5, "candidates": [ 5 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 0, "col": 0, "value": 2, "candidates...
{ "hidden_single": 3, "naked_single": 48 }
sudoku-verify-benchmark-v1
v1
test
test_0008
8
30
9
3
3
[ [ 0, 7, 0, 2, 0, 6, 0, 0, 0 ], [ 0, 8, 0, 5, 0, 9, 0, 3, 6 ], [ 0, 0, 0, 0, 0, 0, 2, 0, 7 ], [ 0, 0, 7, 0, 3, 0, 6, 0, 0 ], [ 3, 0, 0, 0, ...
[ [ 5, 7, 3, 2, 1, 6, 9, 8, 4 ], [ 2, 8, 4, 5, 7, 9, 1, 3, 6 ], [ 6, 9, 1, 3, 4, 8, 2, 5, 7 ], [ 4, 2, 7, 8, 3, 5, 6, 9, 1 ], [ 3, 6, 5, 4, ...
[ { "technique": "naked_single", "row": 5, "col": 0, "value": 9, "candidates": [ 9 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 5, "col": 1, "value": 1, "candidates...
{ "hidden_single": 12, "naked_single": 39 }
sudoku-verify-benchmark-v1
v1
test
test_0009
9
30
9
3
3
[ [ 4, 0, 0, 8, 3, 1, 7, 6, 0 ], [ 0, 6, 0, 0, 7, 0, 9, 0, 8 ], [ 0, 0, 8, 2, 0, 6, 4, 0, 0 ], [ 0, 0, 0, 0, 0, 0, 0, 5, 0 ], [ 0, 1, 0, 0, ...
[ [ 4, 5, 9, 8, 3, 1, 7, 6, 2 ], [ 2, 6, 1, 4, 7, 5, 9, 3, 8 ], [ 7, 3, 8, 2, 9, 6, 4, 1, 5 ], [ 8, 7, 4, 6, 2, 3, 1, 5, 9 ], [ 9, 1, 6, 7, ...
[ { "technique": "naked_single", "row": 5, "col": 6, "value": 8, "candidates": [ 8 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 3, "col": 6, "value": 1, "candidates...
{ "hidden_single": 10, "naked_single": 41 }
sudoku-verify-benchmark-v1
v1
test
test_0010
10
30
9
3
3
[ [ 0, 3, 7, 0, 9, 0, 0, 0, 2 ], [ 2, 0, 8, 4, 0, 0, 0, 0, 0 ], [ 0, 0, 4, 0, 2, 0, 7, 0, 6 ], [ 0, 0, 3, 0, 6, 0, 0, 0, 8 ], [ 0, 0, 0, 0, ...
[ [ 6, 3, 7, 1, 9, 5, 4, 8, 2 ], [ 2, 1, 8, 4, 7, 6, 3, 9, 5 ], [ 9, 5, 4, 3, 2, 8, 7, 1, 6 ], [ 4, 2, 3, 5, 6, 9, 1, 7, 8 ], [ 7, 6, 9, 2, ...
[ { "technique": "hidden_single", "row": 0, "col": 6, "value": 4, "candidates": [ 1, 4, 8 ], "unit_type": "row", "unit_index": 0, "note": "The digit appears in only one cell within the row." }, { "technique": "hidden_single", "row": 1, "col": 4, ...
{ "hidden_single": 10, "naked_single": 41 }
sudoku-verify-benchmark-v1
v1
test
test_0011
11
30
9
3
3
[ [ 3, 7, 0, 0, 6, 1, 0, 0, 0 ], [ 0, 6, 5, 4, 0, 0, 0, 0, 0 ], [ 2, 0, 0, 0, 0, 5, 3, 0, 0 ], [ 0, 5, 0, 0, 3, 0, 1, 0, 0 ], [ 0, 1, 8, 0, ...
[ [ 3, 7, 9, 2, 6, 1, 4, 8, 5 ], [ 8, 6, 5, 4, 7, 3, 2, 9, 1 ], [ 2, 4, 1, 9, 8, 5, 3, 6, 7 ], [ 9, 5, 6, 7, 3, 4, 1, 2, 8 ], [ 7, 1, 8, 5, ...
[ { "technique": "naked_single", "row": 6, "col": 1, "value": 3, "candidates": [ 3 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 6, "col": 2, "value": 7, "candidates...
{ "hidden_single": 9, "naked_single": 42 }
sudoku-verify-benchmark-v1
v1
test
test_0012
12
30
9
3
3
[ [ 7, 5, 0, 8, 9, 2, 0, 0, 0 ], [ 0, 0, 0, 0, 4, 0, 0, 0, 0 ], [ 1, 0, 4, 0, 0, 0, 0, 9, 0 ], [ 0, 0, 1, 0, 0, 5, 0, 8, 9 ], [ 3, 0, 0, 0, ...
[ [ 7, 5, 3, 8, 9, 2, 4, 6, 1 ], [ 8, 9, 2, 1, 4, 6, 5, 3, 7 ], [ 1, 6, 4, 5, 7, 3, 8, 9, 2 ], [ 6, 2, 1, 4, 3, 5, 7, 8, 9 ], [ 3, 4, 8, 9, ...
[ { "technique": "naked_single", "row": 0, "col": 2, "value": 3, "candidates": [ 3 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 0, "col": 8, "value": 1, "candidates...
{ "hidden_single": 9, "naked_single": 42 }
sudoku-verify-benchmark-v1
v1
test
test_0013
13
30
9
3
3
[ [ 0, 9, 0, 6, 0, 8, 0, 0, 0 ], [ 6, 0, 0, 0, 0, 0, 0, 9, 0 ], [ 7, 8, 3, 0, 0, 9, 0, 0, 0 ], [ 3, 4, 7, 0, 0, 0, 2, 0, 5 ], [ 0, 2, 0, 0, ...
[ [ 4, 9, 2, 6, 1, 8, 3, 5, 7 ], [ 6, 1, 5, 7, 4, 3, 8, 9, 2 ], [ 7, 8, 3, 2, 5, 9, 1, 6, 4 ], [ 3, 4, 7, 8, 9, 6, 2, 1, 5 ], [ 8, 2, 1, 3, ...
[ { "technique": "naked_single", "row": 1, "col": 1, "value": 1, "candidates": [ 1 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 3, "col": 5, "value": 6, "candidates...
{ "hidden_single": 2, "naked_single": 49 }
sudoku-verify-benchmark-v1
v1
test
test_0014
14
30
9
3
3
[ [ 0, 0, 0, 0, 5, 0, 0, 1, 0 ], [ 2, 3, 0, 0, 6, 0, 8, 5, 4 ], [ 0, 0, 0, 0, 0, 4, 0, 0, 0 ], [ 0, 0, 0, 0, 9, 6, 5, 7, 0 ], [ 0, 8, 9, 0, ...
[ [ 9, 7, 4, 8, 5, 3, 6, 1, 2 ], [ 2, 3, 1, 9, 6, 7, 8, 5, 4 ], [ 8, 6, 5, 2, 1, 4, 3, 9, 7 ], [ 1, 2, 3, 4, 9, 6, 5, 7, 8 ], [ 4, 8, 9, 7, ...
[ { "technique": "naked_single", "row": 4, "col": 7, "value": 6, "candidates": [ 6 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 4, "col": 8, "value": 1, "candidates...
{ "hidden_single": 13, "naked_single": 38 }
sudoku-verify-benchmark-v1
v1
test
test_0015
15
30
9
3
3
[ [ 0, 5, 1, 0, 3, 0, 0, 7, 0 ], [ 0, 0, 0, 0, 4, 1, 0, 0, 0 ], [ 4, 0, 8, 0, 5, 0, 0, 0, 1 ], [ 0, 0, 0, 1, 0, 0, 7, 0, 2 ], [ 6, 0, 0, 0, ...
[ [ 9, 5, 1, 2, 3, 8, 4, 7, 6 ], [ 2, 6, 3, 7, 4, 1, 8, 5, 9 ], [ 4, 7, 8, 9, 5, 6, 2, 3, 1 ], [ 3, 8, 9, 1, 6, 5, 7, 4, 2 ], [ 6, 4, 5, 3, ...
[ { "technique": "hidden_single", "row": 0, "col": 6, "value": 4, "candidates": [ 2, 4, 6, 8 ], "unit_type": "row", "unit_index": 0, "note": "The digit appears in only one cell within the row." }, { "technique": "hidden_single", "row": 4, "co...
{ "hidden_single": 12, "naked_single": 39 }
sudoku-verify-benchmark-v1
v1
test
test_0016
16
30
9
3
3
[ [ 9, 4, 0, 0, 0, 0, 0, 6, 0 ], [ 6, 0, 3, 0, 0, 1, 0, 5, 0 ], [ 7, 0, 0, 0, 0, 0, 0, 0, 0 ], [ 0, 6, 4, 1, 0, 0, 9, 0, 0 ], [ 8, 3, 9, 5, ...
[ [ 9, 4, 2, 3, 5, 7, 8, 6, 1 ], [ 6, 8, 3, 9, 4, 1, 2, 5, 7 ], [ 7, 5, 1, 2, 8, 6, 3, 9, 4 ], [ 2, 6, 4, 1, 7, 3, 9, 8, 5 ], [ 8, 3, 9, 5, ...
[ { "technique": "naked_single", "row": 3, "col": 7, "value": 8, "candidates": [ 8 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 3, "col": 8, "value": 5, "candidates...
{ "hidden_single": 7, "naked_single": 44 }
sudoku-verify-benchmark-v1
v1
test
test_0017
17
30
9
3
3
[ [ 0, 0, 0, 5, 0, 0, 0, 0, 9 ], [ 7, 0, 0, 3, 0, 1, 0, 0, 2 ], [ 0, 1, 0, 0, 9, 0, 0, 0, 0 ], [ 0, 0, 0, 0, 1, 5, 0, 2, 3 ], [ 0, 2, 7, 0, ...
[ [ 4, 8, 6, 5, 2, 7, 1, 3, 9 ], [ 7, 5, 9, 3, 8, 1, 6, 4, 2 ], [ 2, 1, 3, 4, 9, 6, 8, 5, 7 ], [ 6, 9, 8, 7, 1, 5, 4, 2, 3 ], [ 3, 2, 7, 6, ...
[ { "technique": "naked_single", "row": 0, "col": 1, "value": 8, "candidates": [ 8 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 3, "col": 1, "value": 9, "candidates...
{ "hidden_single": 2, "naked_single": 49 }
sudoku-verify-benchmark-v1
v1
test
test_0018
18
30
9
3
3
[ [ 5, 0, 0, 0, 0, 0, 0, 9, 0 ], [ 0, 3, 0, 0, 0, 0, 8, 0, 0 ], [ 0, 0, 0, 0, 2, 7, 0, 5, 0 ], [ 0, 0, 0, 1, 0, 9, 5, 0, 0 ], [ 7, 9, 0, 0, ...
[ [ 5, 6, 2, 4, 1, 8, 7, 9, 3 ], [ 1, 3, 7, 6, 9, 5, 8, 4, 2 ], [ 9, 8, 4, 3, 2, 7, 1, 5, 6 ], [ 4, 2, 3, 1, 6, 9, 5, 7, 8 ], [ 7, 9, 5, 2, ...
[ { "technique": "naked_single", "row": 6, "col": 2, "value": 1, "candidates": [ 1 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 6, "col": 0, "value": 6, "candidates...
{ "hidden_single": 5, "naked_single": 46 }
sudoku-verify-benchmark-v1
v1
test
test_0019
19
30
9
3
3
[ [ 0, 7, 0, 0, 4, 3, 0, 1, 0 ], [ 0, 0, 0, 2, 7, 0, 9, 0, 0 ], [ 8, 4, 0, 0, 0, 0, 0, 7, 0 ], [ 0, 0, 0, 0, 0, 0, 2, 5, 0 ], [ 3, 6, 0, 1, ...
[ [ 9, 7, 2, 5, 4, 3, 8, 1, 6 ], [ 1, 5, 6, 2, 7, 8, 9, 3, 4 ], [ 8, 4, 3, 9, 6, 1, 5, 7, 2 ], [ 4, 8, 1, 6, 3, 7, 2, 5, 9 ], [ 3, 6, 9, 1, ...
[ { "technique": "naked_single", "row": 4, "col": 4, "value": 5, "candidates": [ 5 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 5, "col": 3, "value": 8, "candidates...
{ "naked_single": 51 }
sudoku-verify-benchmark-v1
v1
test
test_0020
20
30
9
3
3
[ [ 0, 0, 3, 7, 0, 0, 0, 0, 0 ], [ 9, 0, 0, 8, 3, 0, 0, 4, 0 ], [ 0, 7, 8, 2, 0, 0, 0, 9, 6 ], [ 7, 0, 0, 0, 0, 0, 4, 0, 0 ], [ 0, 0, 0, 0, ...
[ [ 2, 4, 3, 7, 6, 9, 1, 5, 8 ], [ 9, 5, 6, 8, 3, 1, 7, 4, 2 ], [ 1, 7, 8, 2, 4, 5, 3, 9, 6 ], [ 7, 9, 5, 1, 2, 6, 4, 8, 3 ], [ 6, 3, 2, 4, ...
[ { "technique": "naked_single", "row": 2, "col": 4, "value": 4, "candidates": [ 4 ], "unit_type": null, "unit_index": null, "note": "Only one candidate remains in the cell." }, { "technique": "naked_single", "row": 5, "col": 4, "value": 9, "candidates...
{ "naked_single": 51 }
sudoku-verify-benchmark-v1
v1
test
test_0021
21
30
9
3
3
[ [ 5, 9, 0, 0, 0, 0, 0, 4, 2 ], [ 0, 0, 0, 0, 0, 8, 0, 0, 0 ], [ 4, 0, 0, 7, 0, 9, 0, 0, 0 ], [ 1, 5, 0, 8, 0, 0, 0, 0, 0 ], [ 2, 0, 0, 0, ...
[ [ 5, 9, 7, 6, 3, 1, 8, 4, 2 ], [ 3, 1, 6, 2, 4, 8, 7, 5, 9 ], [ 4, 8, 2, 7, 5, 9, 1, 3, 6 ], [ 1, 5, 4, 8, 2, 6, 3, 9, 7 ], [ 2, 7, 9, 4, ...
[ { "technique": "hidden_single", "row": 0, "col": 6, "value": 8, "candidates": [ 1, 3, 6, 7, 8 ], "unit_type": "row", "unit_index": 0, "note": "The digit appears in only one cell within the row." }, { "technique": "hidden_single", "row": 0...
{ "hidden_single": 9, "naked_single": 42 }
End of preview. Expand in Data Studio

SudokuJustify

SudokuJustify is a benchmark for evaluating whether a model can produce stepwise, machine-checkable justifications for simple Sudoku reasoning. Each instance contains a 9x9 puzzle, its full solution, and a canonical symbolic trace using a restricted rule set:

  • naked_single
  • hidden_single

The benchmark is designed for evaluation and analysis rather than training. The restricted rule set keeps every claimed step easy to verify automatically.

Dataset Layout

The repository contains four released splits:

  • v1/test.jsonl: main frozen evaluation split.
  • v2/test.jsonl: second frozen evaluation split.
  • v3/test.jsonl: third frozen evaluation split.
  • v1_balanced_attribution/test.jsonl: balanced HS/NS probe sampled from v1 for attribution and control experiments.

Files under v1/, v2/, and v3/ also include metadata.json, which records the generation parameters for each split.

Main Record Schema

Each JSONL record in v1, v2, and v3 contains:

  • benchmark_format_version: dataset schema version string.
  • dataset_id: split identifier.
  • split: split name.
  • sample_id: per-instance identifier.
  • seed: puzzle generation seed.
  • clues: number of givens in the initial board.
  • size: Sudoku side length. This release uses 9.
  • box_rows, box_cols: subgrid shape metadata. This release uses 3 and 3.
  • puzzle: 9x9 integer grid with 0 for empty cells.
  • solution: solved 9x9 integer grid.
  • canonical_steps: canonical symbolic trace. Each step includes:
    • technique
    • row
    • col
    • value
    • candidates
    • unit_type
    • unit_index
    • note
  • difficulty_signature: counts of supported techniques in the canonical trace.

Balanced Attribution Split Schema

Each record in v1_balanced_attribution/test.jsonl corresponds to one intermediate board state drawn from v1. Key fields include:

  • state_id: unique state identifier.
  • source_sample_id: source puzzle from v1.
  • step_idx: canonical step index.
  • board: partially filled 9x9 board.
  • proposal: target fill proposal with row, column, value, and candidate set.
  • gold_technique: gold label (naked_single or hidden_single).
  • gold_unit_type, gold_unit_index: supporting unit for hidden-single cases.
  • valid_hidden_units: valid supporting units for the hidden-single justification.
  • fill_ratio: proportion of filled cells in the board state.

Intended Use

This dataset is intended for:

  • evaluating stepwise reasoning outputs on Sudoku;
  • testing whether generated justifications match board state constraints;
  • probing whether model behavior changes under relabeling or balanced HS/NS controls.

It is not intended as a broad measure of general reasoning.

Limitations

  • The benchmark only covers 9x9 Sudoku.
  • The symbolic rule set is intentionally narrow and only includes naked_single and hidden_single.
  • The balanced attribution split is a probe set, not a main benchmark split.
  • Because the dataset is generated through a constrained symbolic pipeline, it reflects the distribution induced by that pipeline rather than the full space of human Sudoku play.

Synthetic Data

This dataset is synthetic. Puzzles and canonical traces are generated and filtered programmatically.

Downloads last month
92