flowchart LR
A["Tous les hommes sont mortels"] --> P{"Déduction"}
B["Socrate est un homme"] --> P
P --> C["Socrate est mortel"]
L’IA moderne ne sait pas résoudre le Sudoku : analyse des limites de l’induction statistique et de l’avenir neuro-symbolique
mardi 12 mai 2026

De 1950 à 1980, le paradigme symbolique domine : l’intelligence est vue comme une manipulation de symboles et de règles
À la fin des années 1980, le connexionnisme prend de l’ampleur : inspiration depuis les réseaux de neurones du cerveau humain
L’IA neuro-symbolique apparaît ensuite comme une tentative de combiner apprentissage statistique et raisonnement déductif.
La déduction va du général au particulier : si les prémisses sont vraies, la conclusion l’est nécessairement.
flowchart LR
A["Tous les hommes sont mortels"] --> P{"Déduction"}
B["Socrate est un homme"] --> P
P --> C["Socrate est mortel"]
L’induction va du particulier au général : elle propose une règle, sans certitude absolue.
flowchart LR
A["Socrate est mortel"] --> P{"Induction"}
B["Platon est mortel"] --> P
C["..."] --> P
D["Andreas est mortel"] --> P
P --> E["Tous les hommes sont mortels"]
L’abduction part d’un fait surprenant pour proposer l’hypothèse la plus plausible.
flowchart LR
A["Tous les hommes sont mortels"] --> P{"Abduction"}
B["Socrate est mortel"] --> P
P --> C["Socrate est un homme"]
L’application des précédents termes à notre problématique :
Le Sudoku classique consiste à remplir une grille \(9×9\) pour que chaque ligne (Figure 1), colonne (Figure 2) et bloc \(3×3\) (Figure 3) contienne les chiffres de \(1\) à \(9\).
2 1 8 | 3 9 4 | 6 7 5
. . . | . . . | . . .
. . . | . . . | . . .
------+-------+------
. . . | . . . | . . .
. . . | . . . | . . .
. . . | . . . | . . .
------+-------+------
. . . | . . . | . . .
. . . | . . . | . . .
. . . | . . . | . . .
2 . . | . . . | . . .
9 . . | . . . | . . .
4 . . | . . . | . . .
------+-------+------
6 . . | . . . | . . .
7 . . | . . . | . . .
8 . . | . . . | . . .
------+-------+------
1 . . | . . . | . . .
3 . . | . . . | . . .
5 . . | . . . | . . .
2 1 8 | . . . | . . .
9 7 3 | . . . | . . .
4 6 5 | . . . | . . .
------+-------+------
. . . | . . . | . . .
. . . | . . . | . . .
. . . | . . . | . . .
------+-------+------
. . . | . . . | . . .
. . . | . . . | . . .
. . . | . . . | . . .
Résoudre une grille donnée n’implique pas savoir résoudre génériquement toutes les instances du problème.
. 1 8 | . . . | . 7 .
. 7 . | . . 1 | 9 . .
. 6 . | 8 5 . | 1 2 .
------+-------+------
6 . . | 7 . . | 3 . .
7 . . | 5 1 . | . 8 .
8 . 4 | . . 9 | 7 . 5
------+-------+------
. 4 7 | . 9 8 | . 5 .
. . 2 | 6 . 5 | . 3 .
. . 6 | . . . | 2 4 .
Instance donnée aux LLM.
5 1 8 | 9 2 4 | 6 7 3
2 7 3 | 4 6 1 | 9 0 8 🛑
. . . | . . . | . . .
------+-------+------
. . . | . . . | . . .
. . . | . . . | . . .
. . . | . . . | . . .
------+-------+------
. . . | . . . | . . .
. . . | . . . | . . .
. . . | . . . | . . .
ChatGPT produit aussi une réponse incorrecte et finit par interrompre la complétion de la grille.
2 1 8 | 9 6 3 | 5 7 4
5 7 3 | 2 4 1 | 9 6 8
4 6 9 | 8 5 7 | 1 2 3
------+-------+------
6 5 1 | 7 8 4 | 3 9 2
7 9 3 | 5 1 2 | 4 8 6
8 2 4 | 3 6 9 | 7 1 5
------+-------+------
3 4 7 | 1 9 8 | 6 5 2
1 8 2 | 6 7 5 | 4 3 9
9 3 6 | 4 2 1 | 2 4 7 ‼️
Gemini 3 Rapide renvoie une grille erronée, avec une incohérence dans la dernière ligne.
4 1 8 | 9 2 6 | 5 7 3
2 7 5 | 4 3 1 | 9 6 8
9 6 3 | 8 5 7 | 1 2 4
------+-------+------
6 5 1 | 7 8 4 | 3 9 2
7 3 9 | 5 1 2 | 4 8 6
8 2 4 | 3 6 9 | 7 1 5
------+-------+------
3 4 7 | 2 9 8 | 6 5 1
1 9 2 | 6 4 5 | 8 3 7
5 8 6 | 1 7 3 | 2 4 9 ✅
Gemini 3 Pro réussit l’instance simple, grâce à un raisonnement étendu type chaîne de pensée.
Ce test montre qu’un modèle plus puissant peut réussir un cas simple, sans pour autant disposer d’un solveur logique robuste.
Résolution d’un Sudoku difficile par Gemini 3 pro (raisonnement) :
. . . | . 3 . | . . 6
. 1 . | . 5 . | . . .
. . . | . . 9 | 4 . 3
------+-------+------
8 . . | . . . | 5 . .
. . 4 | . . 2 | 1 . 7
6 . 5 | . . . | . . 4
------+-------+------
1 . 9 | 3 . . | . . .
. . . | . 4 . | . 3 .
4 . 2 | . 8 1 | . . .
Instance donnée au LLM.
7 4 8 | 2 3 1 | 9 5 6
2 1 3 | 8 5 6 | 7 4 9
5 9 6 | 7 2 9 | 4 1 3
------+-------+------
8 2 1 | 4 6 7 | 5 9 3
3 9 4 | 5 1 2 | 6* 8 7
6 7 5 | 9 8 3 | 1 2 4
------+-------+------
1 5 9 | 3 7 8 | 2 6 4
9 8 7 | 6 4 5 | 3 1 2
4 3 2 | 1 9 4 | 8 7 5
* Valeur initiale modifiée.
Initiating Grid Analysis
I've begun analyzing the Sudoku grid, identifying the user's objective and input format in French.
Defining Solution Strategy
I've determined text output is the best approach, and I'm starting to build a solving algorithm.
Analyzing Number Placement
I've just made significant deductions on number placement, specifically for '4' and '1'.
I've determined that R7C8 must be '4', and further analyzed Box 9 after.
Analyzing Number Placement (Again)
I've successfully placed '1' at R4C3, and initiated further deductions about other numbers.
Deducing Candidate Values
I've determined some missing numbers and narrowed potential candidates within specific rows and boxes.
Analyzing Candidate Constraints
I've determined possible values and exclusions for Row 7, and reexamined column and box 6 for restrictions.
Analyzing Number Placement (Again)
I've determined some new possible values, applying deductions across rows, columns, and boxes.
Deducing Possible Values
I've determined possible locations for '7', '8', and missing values in Row 5, and I'm examining Box 5.
Analyzing Number Placement (Again)
I've determined R5C4 must be '5' using column constraints and deduced candidates for row and box 5.
Analyzing Number Placement (Again)
I've determined some new possible locations for numbers, revisiting existing areas of interest, like Boxes 2, 4 and 7.
Confirming Number Placement
I've successfully placed '4' at R1C2 using box and column constraints, and I'm further examining 4's grid placement.
Refining Value Placement
I've determined possible candidate locations for '1' in Boxes 3 and 9, based on exclusions.
Confirming New Placement
I've determined R8C9 is '1', which I verified and used to begin Box 3 deductions.
Deducing New Values
I've determined R6C7 is '3', and I'm updating Row 6, finding the placement for missing numbers.
Analyzing Pair Deduction
I've determined a strong candidate pair in Box 4, using row and column analysis.
I've deduced that R5C1 and R5C2 are {3, 9} and R4C2 and R6C2 are {2, 7} within Box 4, due to missing values.
Confirming Box Values
I've determined that R5C5 is '6' and R5C8 is '8', and I'm updating deductions.
Refining Grid Analysis
I've determined possible locations for 5 and 9 within Box 9, using existing constraints.
Refining Candidate Pairs
I've determined a hidden pair of {1, 5} exists at R1C8 and R3C8 within Box 3, by exclusions and constraints.
Refining Row Analysis
I'm now focusing on Box 3, specifically on constraints within Rows 1, 2, and 3, to place '8'.
Confirming Col Deduction
I've determined R9C2 is '3', R5C2 is '9', and R5C1 is '3', significantly progressing column and box analysis.
Refining Placement Analysis
I've determined possible locations for 9 within column 7, and reexamined column and box constraints.
\n\n
Reassessing Row Analysis
I've narrowed down candidate values for Col 1 and reviewed locations of missing numbers in Box 7.
Deducing New Placements
I've determined possible locations for missing values within Columns 3 and 5, based on current exclusions.
Confirming Number Placement
I've determined that R4C6 is '3', R4C4 is '4', and reexamined candidate placements within Box 5.
Chaîne de pensée Gemini
Ainsi, même un modèle plus puissant échoue à résoudre une instance plus compliquée du Sudoku.
L’étude de (Seely et al. 2025) teste 100 Sudokus de tailles 4×4, 6×6 et 9×9 avec plusieurs modèles :
| Modèle (LLM) | \(4×4\) | \(6×6\) | \(9×9\) |
|---|---|---|---|
| o3‑mini‑high | \(73.3 \%\) | \(6.7 \%\) | \(2.9 \%\) |
| Gemini 2.5 Pro | \(60.0 \%\) | \(13.3 \%\) | \(0.0 \%\) |
| GPT‑4.1 | \(13.3 \%\) | \(0.0 \%\) | \(0.0 \%\) |
| Qwen-* | \(40-53 \%\) | \(0.0 \%\) | \(0.0 \%\) |
Les meilleures performances restent loin d’une fiabilité générale, y compris sur les grilles 4×4.


Le penseur semblant soulagé d’apprendre ces perspectives. Néanmoins, il s’agit d’un penseur pensé et dessiné par une IA. – Gemini/Nano Banana

Cytopia | Cité Internationale Universitaire de Paris