Natural Language Watermarking Based on Syntactic Displacement and Morphological Division
Bibliographic record
Abstract
This paper explores a method for Korean text watermarking based on a linguistic analysis scheme using morphemic and syntactic analysis. In this scheme, a predicate nominal is separated into its nominal and its predicate, and syntactic adverbial is displaced. Korean, as an agglutinative language, provides a good basis for this morpheme-based natural language watermarking because a word consists of several morphemes. A Korean word usually consists of a content morpheme and a function morpheme. However, a predicate nominal is an exception, having two content morphemes-nominal and predicate--and one function morpheme. So, we can divide a predicate nominal into a nominal and a predicate. In addition, we also perform syntax-based watermarking. We displace syntactic adverbials using the characteristic that most languages permit displacement of syntactic adverbials within its clause. Combining these morphemic and syntactic characteristics, we propose a method of language watermarking based on syntactic displacement and morphological division. To make our system more secure, we also include a sentence weight value and encode the weight value with a watermark bit. Our watermarking method doesn't change the meaning of the most marked sentences, and it also ensures the naturalness of the sentences. From the experimental results, we show that the rate of unnatural sentences of marked text is reasonable, and the watermarking capacity is better than previous systems. The coverage of marked sentences is also reasonable. Experimental results also show that the marked text retains the same style, and also has the same information without semantic distortion.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".