Leidos was granted a patent for masked-transformer synthetic data generation.

{
  "signal_id": "6c4e0e8f-8075-51a4-b7b0-8f9adbbde269",
  "signal_type": "patents-company",
  "signal_subtype": "patentGrant",
  "detected_at": "2026-09-23T15:07:26+00:00",
  "company": {
    "name": "Leidos",
    "domain": "leidos.com"
  },
  "data": {
    "cpc": [
      {
        "code": "G06N 3/0475",
        "label": "Computing arrangements based on specific computational models"
      }
    ],
    "tags": [
      "Intellectual Property",
      "Research & Development",
      "Artificial Intelligence",
      "Machine Learning",
      "Data Science"
    ],
    "detail": "In September 2026 Leidos patented a method for creating realistic artificial datasets for AI development, testing, and analysis across data-driven applications.",
    "summary": "Leidos was granted a patent for masked-transformer synthetic data generation.",
    "event_at": "2026-09-22",
    "inventors": [
      {
        "name": "Paul F. Roysdon",
        "status": "resolved",
        "is_primary": true,
        "linkedin_url": "https://www.linkedin.com/in/paul-roysdon-ph-d-7628241b8"
      },
      {
        "name": "Manbir S. Gulati",
        "status": "unresolved",
        "is_primary": false
      }
    ],
    "tech_area": "artificial intelligence / machine learning",
    "event_kind": "grant_date",
    "grant_date": "2026-09-22",
    "source_url": "https://patents.google.com/patent/US12743607",
    "cpc_primary": {
      "code": "G06N 3/0475",
      "label": "Computing arrangements based on specific computational models"
    },
    "patent_title": "System and method for generating synthetic data with masked transformers",
    "patent_number": "US12743607",
    "uspto_assignee": "Leidos Inc.",
    "patent_abstract": "A transformer-based modeling architecture for training a generator to generate synthetic data. The architecture includes embedding models for embedding input data having multiple fields containing real data values and constructing individual embedding matrices for each of the multiple fields, a masking model for producing a set of masked fields in the embedded input dataset and replacing the real data value of each field with a mask token; and a transformer model for predicting the real data value for each mask token in each of the masked fields over multiple iterations.",
    "precision_class": "structural",
    "invention_explanation": "A transformer masks selected fields in real records and learns to predict the missing values, allowing it to produce new records that resemble the original data without copying it directly.",
    "technologies_mentioned": [
      "synthetic data",
      "transformers",
      "machine learning",
      "data masking"
    ]
  }
}

curl https://signals.autobound.ai/v1/signals/6c4e0e8f-8075-51a4-b7b0-8f9adbbde269 \
  -H "X-API-KEY: $AUTOBOUND_API_KEY"

curl -X POST https://signals.autobound.ai/v1/companies/enrich \
  -H "X-API-KEY: $AUTOBOUND_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"domain":"leidos.com","limit":20}'