Skip to content
Alshival.Ai
Sign in/up

Research / Papers & tools

Public data.
Open questions.

We study new ideas and build tools that make complex data easier to investigate. Read our papers, explore the evidence, and follow the sources behind it.

01 / Research paper · October 2026

Learning Reversible Stochastic Tensor Transformations

A Path Toward Quantum-Proof MultiModal Encryption

Reversible neural networks learn stochastic image transformations while preserving recovery of the original pixels. This study explores a foundation for future multimodal privacy systems; cryptographic security remains an open research goal.

Samuel Cavazos · Chief Data Scientist, Alshival.Ai

Read the paper View on GitHub
First page of the research paperRead the full paper 19 pages · PDF ↗

02 Interactive research

NHTSA
Complaint Explorer.

What are drivers reporting? Explore related vehicle complaints across major car companies in a shared 3D map.

  • Move from companies to clusters of related accounts.
  • Search by exact words or similar descriptions.
  • Read the original complaint behind each dot.
Open explorer

Download data on Kaggle ↗

New to the map? The ? Help button walks you through it.

Preview of the complaint atlas, with colored points representing complaints from different car companiesExplore the map
Complaints
—
Companies
—
Model years
2016 onward

Start with the accounts.

These are reports submitted to NHTSA. Similarity helps surface related experiences; it does not establish a defect or measure incident rates.

Refreshed from NHTSA.

Completed updates include new and revised complaints. The explorer explains the layout, clustering, and coverage.

Data source: NHTSA ↗