GERO · Audit digest

Mathematical invariants across AI software.

Small contracts—chain rules, tensor shapes, channel symmetry and execution parity—can expose silent failures that large aggregate scores hide.

Published September 4, 2026 · Xamit Kadirbekov · Reproducible software audit
Attribution boundary. The Samsung derivative cluster was identified in this audit. The Apple MLX, JAX, PyTorch and TensorFlow reports originated with the GitHub users linked below and were independently reproduced.

Five compact cases

Independent reproduction

Apple MLX: one-element strided slice

A plain slice has shape (1,3), while vmap produces (4,2,3); reverse-mode gradients are also scattered to the wrong positions.

Original report by mentria-ai →
New audit cluster

Samsung nntrainer: five unary derivatives

Real calcDerivative calls for sqrt, negative, sine, cosine and tangent fail ordinary chain-rule oracles when the incoming derivative differs from one. A minimal in-place correction makes all five focused regression tests pass.

Project source →
Numerical design question

Google JAX: log2 of exact powers of two

Reproduced 46/254 float32 and 105/254 bfloat16 mismatches. The observation is firm; the boundary between API defect and accuracy/performance tradeoff remains under discussion.

Original report by gonnet →
Independent reproduction

PyTorch: eager versus Inductor

For unrepresentable float32 clamp bounds, eager raises while Inductor silently returns the input. The split reproduces for clamp, clamp_min and clamp_max.

Original report by laolvfan →
Independent reproduction

TensorFlow: MaxPool channel symmetry

Three identical channels produce [NaN, NaN, 3] when packed and [3,3,3] when processed separately. The proposed upstream fix is still open.

Original report by lackjava-creator →

The GERO method

State a necessary invariant. Construct the smallest counterexample. Run the actual library code. Preserve the version, command and raw output. Then distinguish a confirmed defect from a documentation mismatch or a numerical boundary.

This is the product direction of GERO: transform necessary claims into explicit contracts, route them to heterogeneous checks, search for counterexamples and return bounded verdicts with replayable evidence.

Read the full canonical article →