paper: doi:10.1007/s11192-026-05695-x
Data license: Space charter; records cite primary sources · Data source: TeamScience Space repository
This data as json
| lom_id | doi | openalex | s2_paper_id | arxiv | pmid | pmcid | acl | title | year | venue | primary_field | primary_field_id | oa_url | ingested_ts | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| doi:10.1007/s11192-026-05695-x | 10.1007/s11192-026-05695-x | W7168048551 | Exploiting large language models in peer review: indirect prompt injection attacks and integrity probes | 2026 | Scientometrics | Social Sciences | https://openalex.org/fields/33 | https://link.springer.com/content/pdf/10.1007/s11192-026-05695-x.pdf | 2026-09-02T02:37:09Z | openalex |