Tokenized Data Streaming for Multi-Modal AI in Vehicles
Summary
Tokenized data streaming for multi-modal AI in vehicles
What changed
USPTO published patent application US20260097781A1 describing a system for tokenized data streaming between in-vehicle system-on-chips (SoCs) and external AI accelerators in multi-modal language models for vehicles. The system encodes raw sensor data (vision, audio) into tokens on a first device and streams them to a second device hosting the language model inference server, which returns detection results used to control vehicle operations.
Technology and automotive companies developing autonomous driving systems, AI accelerators, or in-vehicle computing platforms should review this application's claims for potential overlap with existing or planned products. Patent monitoring services should track this application's prosecution for any divisional or continuation filings that may expand the scope of protection.
What to do next
- Monitor for updates on patent prosecution status
- Review for potential licensing implications if developing similar AI vehicle technology
Archived snapshot
Apr 9, 2026GovPing captured this document from the original source. If the source has since changed or been removed, this is the text as it existed at that time.
TOKENIZED DATA STREAMING FOR MULTI-MODAL LANGUAGE MODELS
Application US20260097781A1 Kind: A1 Apr 09, 2026
Inventors
Rajath Bellipady Shetty, Niral Lalit Pathak, Ratin Kumar
Abstract
In various examples, a multi-modal language model may be split up and hosted by multiple devices. For example, a modality (e.g., vision, audio) encoder and/or projector of the multi-modal language model (e.g., a vision language model) may be hosted on one device (e.g., an in-vehicle SoC) that encodes raw sensor data into corresponding tokens and streams the tokens to a second device (e.g., an external graphic processing unit (GPU) or artificial intelligence (AI) accelerator) that hosts an inference server and a language model (LM) of the multi-modal language model. The LM may return a response indicating the result(s) of the requested detection task, and the response may be used to take some responsive action (e.g., control one or more operations of an ego-machine).
CPC Classifications
B60W 60/001 H04L 9/3213
Filing Date
2024-10-03
Application No.
18905193
Related changes
Get daily alerts for USPTO Patent Applications - Networking (H04L)
Daily digest delivered to your inbox.
Free. Unsubscribe anytime.
Source
About this page
Every important government, regulator, and court update from around the world. One place. Real-time. Free. Our mission
Source document text, dates, docket IDs, and authority are extracted directly from USPTO.
The plain-English summary, classification, and "what to do next" steps are AI-generated from the original text. Cite the source document, not the AI analysis.
Classification
Who this affects
Taxonomy
Browse Categories
Get alerts for this source
We'll email you when USPTO Patent Applications - Networking (H04L) publishes new changes.