Paper 01 / 2026

Research

Abstract and introduction sections from Kinematic Features Are All You Need

Detecting synthetic mouse trajectories under white-box adversarial optimization.

≤ 0.001 Equal error rate
> 99.5% TPR at FPR < 0.1%
17 Kinematic features

Kinematic detection under adversarial optimization.

Kinematic Features Are All You Need: Detecting Synthetic Mouse Trajectories Under Adversarial Optimization

Modern mouse generators borrow from motor-control science: Fitts' Law timing, lognormal submovements, physiological tremor. The paper asks whether those additions actually hold up when a detector measures the shape of movement instead of how human it looks.

Across the Balabit and BOUN mouse-dynamics datasets, the 17-feature framework reaches an equal error rate at or below 0.001. After one detector retraining round, a white-box attacker's mean evasion score falls from 0.999 to 0.010 by round five.

01 ≤ 0.001 Equal error rate
02 > 99.5% TPR at FPR < 0.1%
03 17 Kinematic features
04 43,216 Human trials
05 29 Users across two datasets

Contributions.

The evaluation indicates that a single parametric generator cannot satisfy every measured human movement constraint at once.

  1. 01

    A detector built only from raw x, y, and time trajectories, with no hardware fingerprinting or spectral shortcuts.

  2. 02

    Five rounds of white-box Bayesian optimization that fail to sustain evasion after one detector retraining round.

  3. 03

    A taxonomy of 32 kinematic features across six families, including 15 polling-rate confounds that should be excluded.

  4. 04

    SigmaDrift, an open-source motor-control generator for future trajectory detection research.

Paper and source.

The PDF, archival record, detector implementation, and generator article are linked below.

Mouse dynamics / adversarial ML / motor control / anti-cheat / behavioral biometrics