Generation of optimized knowledge-based language model through knowledge graph multi-alignment

This invention describes a way to make computer programs better at understanding human language. It does this by teaching a language program and a knowledge program together. The knowledge program learns from two different knowledge maps that have been carefully matched up. The claims specifically narrow this to matching knowledge maps in different languages, so the system can learn about one language from knowledge in another.

Why it matters: Filed just before the explosion of powerful large language models (LLMs), this approach to integrating external knowledge with language understanding has become significantly more feasible and effective. Modern LLMs and advanced cross-lingual techniques make the joint training and multi-language alignment described much more powerful than in 2021.

Status: Not in forceListed as no longer active. The specific reason is not in the record we hold.
How hard to build: SpecializedSoftware logic, AI/ML models, data pipelines

3 AI reinvention directions for this patent

See all three

AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.

Reinvent this with AI

Where this one came from

Every morning we go through millions of abandoned patents and pick out 15 worth your time. This was one of today’s.

This one

+10