The Forgetting Engine benchmark program

2,000 trials

2D Protein Folding

Approximately 80% relative improvement in the stated comparison

Internal benchmark against the documented Monte Carlo baseline

4,000 trials

3D Protein Folding

25.8% success versus 3.9%, approximately 361.8% relative improvement

Largest reported relative gap in this research portfolio

Scale series trials

Traveling Salesman

Larger relative gaps were reported at larger tested instances

Benchmark-specific trend, not a universal scaling law

250 trials

Vehicle Routing

Up to 89.3% improvement at the largest tested scale

Compared with the stated routing baseline and configuration

300 trials

Neural Architecture Search

Reported accuracy gains ranged from 3.8% to 8.4%

Internal search benchmark; external replication remains needed

5,000 trials

Quantum Compilation

27.8% gate reduction and 3.7% fidelity gain were reported

Simulator-based comparison under the documented compilation setup

Important

A 361.8% relative success-rate difference in protein folding is not the same quantity as an 89.3% routing improvement or a 27.8% gate reduction. These numbers should be read within their own experiments, not combined into one universal score.

Complexity inversion is an observed pattern, not a declared law.

In several CONEXUS benchmark series, the relative advantage over the chosen baseline increased at larger tested scales. That is the phenomenon CONEXUS calls complexity inversion. Establishing a general scaling law would require preregistered experiments, stronger competing methods, multiple independent implementations, and replication outside the CONEXUS team.

Observed

Larger relative gaps in selected benchmark series as tested scale increased.

Not yet established

A universal rule that the Forgetting Engine improves with every form of complexity or defeats all conventional algorithms.

VIEW COMPANY SITE