{"paper":{"title":"CitePrism: Human-in-the-Loop AI for Citation Auditing and Editorial Integrity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"CitePrism combines AI analysis with human review to help editors audit citation relevance and integrity.","cross_cats":["cs.AI","cs.DL"],"primary_cat":"cs.SI","authors_text":"Binh Vu, Budanur Madappa Darshan Gowda, Gowrika Mahesh, Kavana Gopladevarahalli Papegowda, Mehrdad Jalali, Prajwal Basavaraj, Swati Chandna","submitted_at":"2026-05-15T14:31:45Z","abstract_excerpt":"Editors and reviewers are expected to ensure that manuscripts cite relevant, accurate, current, and ethically appropriate literature, yet manuscript-level citation auditing remains largely manual, fragmented, and difficult to scale. Citation context, metadata quality, self-citation patterns, and bibliographic integrity all affect whether a reference appropriately supports a local claim. We present CitePrism, a transparent hybrid decision-support framework for editorial citation auditing that combines LLM-assisted contextual reasoning, embedding-based semantic similarity, metadata verification,"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"These results suggest that CitePrism may support conservative editorial screening and citation-quality triage, but they do not establish general editorial performance.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The agreement level and flagging performance observed in the preliminary validation on a single case-study manuscript from pavement engineering will hold for other manuscripts, domains, and annotators.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"CitePrism is a human-in-the-loop AI framework for auditing manuscript citations that demonstrated moderate agreement with human relevance judgments in a pilot validation on one paper.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"CitePrism combines AI analysis with human review to help editors audit citation relevance and integrity.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"dd5ab91936e08329fb33e10d122700df5c0ef5f0d2e338355100dace3ce40f36"},"source":{"id":"2605.16000","kind":"arxiv","version":1},"verdict":{"id":"f03792d2-f77e-4012-a611-f8547d325feb","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T18:56:14.146681Z","strongest_claim":"These results suggest that CitePrism may support conservative editorial screening and citation-quality triage, but they do not establish general editorial performance.","one_line_summary":"CitePrism is a human-in-the-loop AI framework for auditing manuscript citations that demonstrated moderate agreement with human relevance judgments in a pilot validation on one paper.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The agreement level and flagging performance observed in the preliminary validation on a single case-study manuscript from pavement engineering will hold for other manuscripts, domains, and annotators.","pith_extraction_headline":"CitePrism combines AI analysis with human review to help editors audit citation relevance and integrity."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.16000/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T19:01:37.413372Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T19:01:18.995367Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T17:33:42.174987Z","status":"skipped","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T17:01:55.658786Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"1fafc64d93e5a7ede2aa813070c3ea8eaeff65a7578037a75223eeb60d1af888"},"references":{"count":32,"sample":[{"doi":"","year":2020,"title":"M. Biagioli and A. Lippman,Gaming the Metrics: Misconduct and Manipulation in Academic Research. Cambridge, MA: MIT Press, 2020","work_id":"3867c001-0ca3-47ee-85d6-1d6ade137eab","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Citation context analysis as a method for conducting rigorous and impactful literature reviews,","work_id":"52e02a13-1135-4f02-8adc-baf689c5f486","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"COPE guidelines on authorship and contributorship,","work_id":"f936cc6f-19e2-4b38-9a04-cedee6592cb4","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Self-citations: Necessary evil or sign of limited scholar- ship?","work_id":"db978ddd-b0e4-4539-b566-53c2f841242c","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"The global burden of journal peer review and the literature on peer review,","work_id":"0fef435d-01a7-44ae-ae53-d8d4e6a94081","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":32,"snapshot_sha256":"0c4530da9e2874bf548aa37b73131f6a4c42f7715d74a223f17cb1f1dbb7af08","internal_anchors":1},"formal_canon":{"evidence_count":1,"snapshot_sha256":"6238bce96dbed6b753716d6aa791b5b081c952e47cd621b6fbfc61a8a14c8bd7"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}