Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.24259 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908740645552128 |
|---|---|
| author | Rose, Michael E. Ghosh, Mainak Erhardt, Sebastian Li, Cheng Buunk, Erik Harhoff, Dietmar |
| author_facet | Rose, Michael E. Ghosh, Mainak Erhardt, Sebastian Li, Cheng Buunk, Erik Harhoff, Dietmar |
| contents | We develop a language similarity model suitable for working with patents and scientific publications at the same time. In a horse race-style evaluation, we subject eight language (similarity) models to predict credible Patent-Paper Citations. We find that our Pat-SPECTER model performs best, which is the SPECTER2 model fine-tuned on patents. In two real-world scenarios (separating patent-paper-pairs and predicting patent-paper-pairs) we demonstrate the capabilities of the Pat-SPECTER. We finally test the hypothesis that US patents cite papers that are semantically less similar than in other large jurisdictions, which we posit is because of the duty of candor. The model is open for the academic community and practitioners alike. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_24259 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Tracing the Flow of Knowledge From Science to Technology Using Deep Learning Rose, Michael E. Ghosh, Mainak Erhardt, Sebastian Li, Cheng Buunk, Erik Harhoff, Dietmar Computation and Language I.2; J.4 We develop a language similarity model suitable for working with patents and scientific publications at the same time. In a horse race-style evaluation, we subject eight language (similarity) models to predict credible Patent-Paper Citations. We find that our Pat-SPECTER model performs best, which is the SPECTER2 model fine-tuned on patents. In two real-world scenarios (separating patent-paper-pairs and predicting patent-paper-pairs) we demonstrate the capabilities of the Pat-SPECTER. We finally test the hypothesis that US patents cite papers that are semantically less similar than in other large jurisdictions, which we posit is because of the duty of candor. The model is open for the academic community and practitioners alike. |
| title | Tracing the Flow of Knowledge From Science to Technology Using Deep Learning |
| topic | Computation and Language I.2; J.4 |
| url | https://arxiv.org/abs/2512.24259 |