Unintended Negative Impacts of Promotional Language in Patent Evaluation

Bingkun Zhao 1, Chenwei Zhang 2, Hao Peng 1,3∗ · 2026

Read Unintended Negative Impacts of Promotional Language in Patent Evaluation by Bingkun Zhao 1, Chenwei Zhang 2, Hao Peng 1,3∗ (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-04926.txt.

Introduction

Bingkun Zhao, Chenwei Zhang, and Hao Peng study how promotional language affects the evaluation of patent applications in the United States. Hype words are increasingly used to sell new ideas in science, but their role for inventions was unclear. The team built a checked list of one hundred thirty-five promotional words for this study.

Across millions of patent applications, more promotional wording was linked to lower chances of approval, ownership transfer, and successful appeal. Yet such language was not hiding weak inventions, since it tracked real novelty and later citation impact. Tolerance also varied by examiner, with men and experienced examiners accepting promotional framing more readily.

Catalog record

CREATORBingkun Zhao 1, Chenwei Zhang 2, Hao Peng 1,3∗
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE62 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.04926

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Unintended Negative Impacts of Promotional Language in Patent Evaluation Bingkun Zhao 1, Chenwei Zhang 2, Hao Peng 1,3∗ Abstract Promotional language has been increasingly used to aid the communication of innovative ideas in science. Yet, less is known about its role in the context of technological innovation. Here, we use a validated and domain-diagnosed lexicon of 135 promotional words to study the association between promotional language and patent evaluation outcomes among 2.7 million USPTO patent applications. Our large-scale study reveals three unexpected findings. First, in contrast to scientific evaluation, we find that a higher frequency of promotional words is negatively associated with the probability of an application being (i) granted a patent, (ii) transferred ownership, and (iii) successfully appealed. This promotional penalty holds even after accounting for a range of confounding factors and is largely robust across different technological areas. Among matched samples, the difference in the success rate between the lowest and highest promotional density quintile is 5.5, 5.9, and 5.3 percentage points for patentability, transferability, and rejection reversal. Second, contrary to institutional skepticism, we show that promotional language is not a mask of weak technology, but objectively reflects the degree of combinatorial novelty and future citation impact. Third, digging into the mechanisms, we find that the tolerance to promotional framing is strongly moderated by human factors, with men and experienced examiners showing a higher acceptance of promotional narratives than women and novice examiners. By revealing an emerging paradox in the patent system, our study offers theoretical and practical implications for improving patent evaluation through more objective scrutiny of linguistic patterns in patent filings. {affiliations} Department of Data Science, City University of Hong Kong, Hong Kong SAR, China Faculty of Education, The University of Hong Kong, Hong Kong SAR, China Northwestern Institute on Complex Systems, Northwestern University, Evanston, IL, USA ∗ Correspondening author (Email: [email protected]) Introduction Patents are foundational to technological progress and economic growth [ 1 , 2 , 3 ] . They incentivize invention by granting temporary monopoly rights in exchange for public disclosure of novel technical solutions [ 4 , 5 , 6 ] . In the United States alone, the patent office processes hundreds of thousands of applications annually, with examiners at the U.S. Patent and Trademark Office (USPTO) serving as critical gatekeepers who must determine whether an invention meets statutory requirements of novelty, utility, and enablement [ 7 , 8 ] . Successful patents not only validate individual creativity but also shape firm strategy, industry competition, and the broader innovation ecosystem [ 9 , 10 ] . However, the patent examination process hinges almost entirely on linguistic communication: examiners evaluate novelty and market potential primarily through written claims, specifications, and drawings submitted by applicants [ 11 , 12 ] . The clarity, precision, and persuasive framing of these documents directly influence whether truly novel ideas are recognized as such or dismissed as obvious in light of prior art [ 13 , 14 ] . This dependence on language raises a fundamental question in technological innovation management: How can inventors effectively communicate the novelty of their ideas, and what linguistic features influence evaluators’ recognition of originality? Novel ideas are inherently uncertain and difficult to convey [ 15 ] . They often recombine distant knowledge domains in ways that defy easy categorization [ 16 ] , making them vulnerable to under-recognition when description fails to highlight their distinctiveness amid cognitive overload and specialization among evaluators [ 17 , 18 , 19 ] . Linguistic choices, particularly those marketing language that emphasize originality, significance, and transformative potential, may be able to serve as powerful cognitive signals, directing attention to an idea’s merits and bridging the gap between technical substance and evaluative judgment [ 20 , 21 ] . Computational linguistic analyses of scientific text have begun to quantify these dynamics, revealing that subtle intensified lexical patterns can profoundly affect the success of idea adoption even when underlying quality is held constant [ 22 ] . One special class of marketing rhetoric is the use of promotional language, which encompasses value-laden adjectives, such as “groundbreaking,” “transformative,” and “unprecedented,” that amplify claims of originality and impact. Recent work on scientific evaluation of research grants provides compelling evidence for the power of such enthusiastic linguistic signaling. In high-stakes grant evaluation contexts, where reviewers must rapidly assess aspirational proposals under resource constraints, promotional language has emerged as a robust predictor of funding success. For instance, a prior study conducted a large-scale semantic analysis of full-text grant applications from three world-renowned funding agencies [ 20 ] . They found that a higher percentage of promotional words was associated with up to a doubling in the probability of funding, net of other applicants and proposal characteristics. Moreover, promotional language itself was linked with the proposal’s inherent novelty and the citation impact of resulting publications, suggesting that it functions not merely as rhetorical embellishment but as an effective vehicle for communicating scientific merits [ 20 ] . Parallel trends have been found in published papers where promotional language correlates with impact metrics [ 23 , 22 ] . These findings underscore a broader insight: in peer-review systems designed to identify and support frontier science, linguistic framing helps overcome the well-documented “novelty rejection” bias [ 24 , 25 , 26 ] , whereby evaluators favor conventional ideas that are easier to assess and predict. Promotional language appears to help harmonize substance with presentation, making innovative ideas legible and compelling [ 20 ] . The patent system, however, operates under a distinct evaluative logic. Unlike grant review, which rewards aspirational vision and potential societal and academic impact, patent examination is a legal-technical adjudication emphasizing objective prior-art comparisons, enablement, and claim definiteness [ 27 ] . Examiners, typically trained engineers or scientists, are often instructed to apply standardized search protocols and legal precedents, with less room for subjective enthusiasm [ 28 , 29 ] . Promotional language that successfully signals novelty in grant contexts may therefore produce unintended consequences in patent review: it could be interpreted as overstatement, complicating the examiner’s ability to map claims cleanly against prior art or raising concerns about enablement and clarity. Despite rich literature on patent evaluation that has focused on patent quality (e.g., citation-based measures of impact [ 30 , 31 ] ), examiner heterogeneity [ 7 , 8 , 29 ] , and linguistic features of disclosure (such as readability and university vs. corporate patent text [ 14 ] ), empirical research has largely overlooked how promotional language specifically influences patent outcomes. Our study fills this critical gap. Leveraging computational linguistics on a large corpus of nearly 3 million patent applications from USPTO, we examine the prevalence and effects of promotional language in different patent outcomes. Consistent with its role in signaling novelty elsewhere in science, we find that promotional language is positively associated with objective measures of an application’s underlying novelty (i.e., recombination-based metrics and post-grant citation patterns). Yet it is robustly linked to higher rejection rates by examiners and the market, creating unintended negative consequences for the very inventors whose ideas are more innovative. These results illuminate a previously unrecognized tension in the patent system: linguistic tools that aid recognition of novelty in one evaluative arena can undermine

Read the complete text