Preference Policy Optimization Patents: Leaders & Trends 2026
Preference Based Policy Optimization Patents: Who Leads and Where the Gaps Sit
37.2% concentration at the top. The five most active filers account for 633 of the 1,702 records in scope, but the remaining ranked leaders thin out fast into single- and double-digit filers. Filings moved from 66 in 2017 to a peak of 318 in 2023, with the last complete year (2024) at 201 — up 44% from 2021's 140. Treat 2025 and 2026 figures as provisional given the typical 18-month publication lag.
- 1MOBILEYE VISION TECH LTD15.0%
- 2STRONG FORCE VCN PORTFOLIO 2019 LLC9.2%
- 3GOOGLE LLC5.8%
- 4STATE FARM MUTAL AUTOMOBILE INSURANCE COMPANY4.0%
- 5STRONG FORCE TX PORTFOLIO 2018 LLC3.2%
See the full preference based policy optimization analysis in Eureka
- The complete ranking, not just the top five
- Every IPC branch with its share of the corpus
- The most-cited records, and where claim space is still thin
Common questions about this landscape
How many patents exist for preference-based policy optimization?
This dataset covers 1,702 published records filed or published between 2015 and mid-2026. The count spans reward model training, policy gradient methods, pairwise preference data handling and related reward-shaping and KL-regularization techniques. Because publication lags filing by roughly 18 months, the true count for 2025 and 2026 filings will rise as those applications publish, so the current total should be read as a floor rather than a ceiling.
Who are the leading filers in this space?
Filing is concentrated at the top: the leading assignee holds 256 records, and the five most active filers combine for 633 records, or 37.2% of the 1,702 in scope. Past the top ten filers, who together hold 47.1% of the field, activity drops quickly into single-digit filers, indicating a genuine long tail rather than a handful of firms controlling the entire space. The mix of leaders includes automotive, robotics-platform and large technology-licensing firms rather than a single industry vertical.
Is patent filing in preference-based policy optimization growing or slowing?
Filing grew 44% from 140 records in 2021 to 201 in 2024, with an interim peak of 318 records in 2023 — the most complete evidence of trend direction available. Raw counts for 2025 and 2026 appear lower, but that reflects the normal 18-month lag between filing and publication, not an actual contraction. Any claim that the field is slowing should be checked against complete-year data only, which currently means 2024 or earlier.
Disclaimer. This analysis is based on Patsnap Eureka data drawn from a limited snapshot of global patent records and is provided for general information and reference only. Patent data carries inherent limitations — recent filings are under-counted because of publication lag, counts may be on a record or family basis, classification and applicant-name data may contain errors or duplicates, and the underlying search query defines the scope shown — so the analysis may be incomplete or inaccurate and may not reflect the full technology landscape.
Nothing here is an exhaustive prior-art, novelty, freedom-to-operate or validity search, nor does it constitute legal, financial or professional advice, and it should not be relied upon as such. Verify independently and review with qualified patent and legal professionals before acting on it.
Method: Filing trend and technology composition. Derived from a Patsnap search on Preference Based Policy Optimization covering 2015–2026, data cut-off 2026-07-31. Counts reflect published records only and shift as new filings publish. Every share divides by all records in scope. Data: Patsnap Eureka. See the full landscape report.