Understand the failure.
Verify the repair.
Small, reproducible software failures. The broken implementation, the fix that didn’t work, and the one that passed—preserved together.
Explore the cases ↓How results are verified ↗WHAT THE ARCHIVE CONTAINS
100840 executable cases. 20168 are open.
Every case records the implementation that fails, the fix that did not work, and the repair that passed its checks—with recorded outputs and source hashes. This release adds 100840 cases across 20168 failure mechanisms and 254 domains.
The open tier gives you the failure and the unsuccessful fix for one case in every mechanism. The remaining 80672 cases, 5 variants per mechanism, are member-only: the verified repair, its recorded checks, and the full fixture suite are held in the member archive. Read the methodology ↗
A RECORD OF WHAT WENT WRONG
Browse the archive / 100840
Vector dot product · case 01
The product of sums introduces cross-coordinate products.
Vector dot product · case 02
The product of sums introduces cross-coordinate products.
Vector dot product · case 03
The product of sums introduces cross-coordinate products.
Vector dot product · case 04
The product of sums introduces cross-coordinate products.
Vector dot product · case 05
The product of sums introduces cross-coordinate products.
Vector hadamard product · case 01
Elementwise addition replaces multiplication.
Vector hadamard product · case 02
Elementwise addition replaces multiplication.
Vector hadamard product · case 03
Elementwise addition replaces multiplication.
Vector hadamard product · case 04
Elementwise addition replaces multiplication.
Vector hadamard product · case 05
Elementwise addition replaces multiplication.
Three dimensional cross product · case 01
Symmetric sums replace antisymmetric determinant terms.
Three dimensional cross product · case 02
Symmetric sums replace antisymmetric determinant terms.
Three dimensional cross product · case 03
Symmetric sums replace antisymmetric determinant terms.
Three dimensional cross product · case 04
Symmetric sums replace antisymmetric determinant terms.
Three dimensional cross product · case 05
Symmetric sums replace antisymmetric determinant terms.
Matrix two determinant · case 01
The off-diagonal product is added instead of subtracted.
Matrix two determinant · case 02
The off-diagonal product is added instead of subtracted.
Matrix two determinant · case 03
The off-diagonal product is added instead of subtracted.
Matrix two determinant · case 04
The off-diagonal product is added instead of subtracted.
Matrix two determinant · case 05
The off-diagonal product is added instead of subtracted.
Matrix trace · case 01
Every matrix entry is summed rather than only the diagonal.
Matrix trace · case 02
Every matrix entry is summed rather than only the diagonal.
Matrix trace · case 03
Every matrix entry is summed rather than only the diagonal.
Matrix trace · case 04
Every matrix entry is summed rather than only the diagonal.
Matrix trace · case 05
Every matrix entry is summed rather than only the diagonal.
Matrix transpose rectangular · case 01
Copying rows leaves row and column axes unchanged.
Matrix transpose rectangular · case 02
Copying rows leaves row and column axes unchanged.
Matrix transpose rectangular · case 03
Copying rows leaves row and column axes unchanged.
Matrix transpose rectangular · case 04
Copying rows leaves row and column axes unchanged.
Matrix transpose rectangular · case 05
Copying rows leaves row and column axes unchanged.
Matrix vector product · case 01
Products between nonmatching coordinates contaminate each row.
Matrix vector product · case 02
Products between nonmatching coordinates contaminate each row.
Matrix vector product · case 03
Products between nonmatching coordinates contaminate each row.
Matrix vector product · case 04
Products between nonmatching coordinates contaminate each row.
Matrix vector product · case 05
Products between nonmatching coordinates contaminate each row.
Matrix multiply row column · case 01
Rows of the right operand are used where columns are required.
Matrix multiply row column · case 02
Rows of the right operand are used where columns are required.
Matrix multiply row column · case 03
Rows of the right operand are used where columns are required.
Matrix multiply row column · case 04
Rows of the right operand are used where columns are required.
Matrix multiply row column · case 05
Rows of the right operand are used where columns are required.
Outer product shape · case 01
The outer product axes are reversed.
Outer product shape · case 02
The outer product axes are reversed.
Outer product shape · case 03
The outer product axes are reversed.
Outer product shape · case 04
The outer product axes are reversed.
Outer product shape · case 05
The outer product axes are reversed.
Matrix frobenius norm squared · case 01
Squaring a total introduces cancellation and cross terms.
Matrix frobenius norm squared · case 02
Squaring a total introduces cancellation and cross terms.
Matrix frobenius norm squared · case 03
Squaring a total introduces cancellation and cross terms.
Matrix frobenius norm squared · case 04
Squaring a total introduces cancellation and cross terms.
Matrix frobenius norm squared · case 05
Squaring a total introduces cancellation and cross terms.
Matrix one norm · case 01
Maximum row sum computes the infinity norm.
Matrix one norm · case 02
Maximum row sum computes the infinity norm.
Matrix one norm · case 03
Maximum row sum computes the infinity norm.
Matrix one norm · case 04
Maximum row sum computes the infinity norm.
Matrix one norm · case 05
Maximum row sum computes the infinity norm.
Matrix infinity norm · case 01
Maximum column sum computes the one-norm.
Matrix infinity norm · case 02
Maximum column sum computes the one-norm.
Matrix infinity norm · case 03
Maximum column sum computes the one-norm.
Matrix infinity norm · case 04
Maximum column sum computes the one-norm.
Matrix infinity norm · case 05
Maximum column sum computes the one-norm.
Two by two adjugate · case 01
Off-diagonal cofactors lack negative signs.
Two by two adjugate · case 02
Off-diagonal cofactors lack negative signs.
Two by two adjugate · case 03
Off-diagonal cofactors lack negative signs.
Two by two adjugate · case 04
Off-diagonal cofactors lack negative signs.
Two by two adjugate · case 05
Off-diagonal cofactors lack negative signs.
Two by two inverse rational · case 01
The adjugate is not divided by the determinant.
Two by two inverse rational · case 02
The adjugate is not divided by the determinant.
Two by two inverse rational · case 03
The adjugate is not divided by the determinant.
Two by two inverse rational · case 04
The adjugate is not divided by the determinant.
Two by two inverse rational · case 05
The adjugate is not divided by the determinant.
Matrix symmetry predicate · case 01
Diagonal positivity is unrelated to transpose symmetry.
Matrix symmetry predicate · case 02
Diagonal positivity is unrelated to transpose symmetry.
Matrix symmetry predicate · case 03
Diagonal positivity is unrelated to transpose symmetry.
Matrix symmetry predicate · case 04
Diagonal positivity is unrelated to transpose symmetry.
Matrix symmetry predicate · case 05
Diagonal positivity is unrelated to transpose symmetry.
Matrix diagonal predicate · case 01
Nonzero diagonal entries do not constrain off-diagonal entries.
Matrix diagonal predicate · case 02
Nonzero diagonal entries do not constrain off-diagonal entries.
Matrix diagonal predicate · case 03
Nonzero diagonal entries do not constrain off-diagonal entries.
Matrix diagonal predicate · case 04
Nonzero diagonal entries do not constrain off-diagonal entries.
Matrix diagonal predicate · case 05
Nonzero diagonal entries do not constrain off-diagonal entries.
Two by two characteristic coefficients · case 01
The trace coefficient has the wrong sign.
Two by two characteristic coefficients · case 02
The trace coefficient has the wrong sign.
Two by two characteristic coefficients · case 03
The trace coefficient has the wrong sign.
Two by two characteristic coefficients · case 04
The trace coefficient has the wrong sign.
Two by two characteristic coefficients · case 05
The trace coefficient has the wrong sign.
Orthogonality zero dot · case 01
Zero coordinate sum is confused with zero mutual dot product.
Orthogonality zero dot · case 02
Zero coordinate sum is confused with zero mutual dot product.
Orthogonality zero dot · case 03
Zero coordinate sum is confused with zero mutual dot product.
Orthogonality zero dot · case 04
Zero coordinate sum is confused with zero mutual dot product.
Orthogonality zero dot · case 05
Zero coordinate sum is confused with zero mutual dot product.
Two dimensional affine transform · case 01
The homogeneous translation column is omitted.
Two dimensional affine transform · case 02
The homogeneous translation column is omitted.
Two dimensional affine transform · case 03
The homogeneous translation column is omitted.
Two dimensional affine transform · case 04
The homogeneous translation column is omitted.
Two dimensional affine transform · case 05
The homogeneous translation column is omitted.
Three dimensional scalar triple product · case 01
Coordinatewise triple multiplication omits determinant permutations.
Three dimensional scalar triple product · case 02
Coordinatewise triple multiplication omits determinant permutations.
Three dimensional scalar triple product · case 03
Coordinatewise triple multiplication omits determinant permutations.
Three dimensional scalar triple product · case 04
Coordinatewise triple multiplication omits determinant permutations.
Three dimensional scalar triple product · case 05
Coordinatewise triple multiplication omits determinant permutations.
INSPECTABLE BY DESIGN
Every result has a runnable source.
Runnable implementations with recorded outputs, source hashes, and explicit contracts. Related variants share a failure mechanism and belong together in evaluation splits.
Read the methodology ↗