{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NFUDUQHTQJCZQNSTYEVM6B3RLI","short_pith_number":"pith:NFUDUQHT","schema_version":"1.0","canonical_sha256":"69683a40f38245983653c12acf07715a2fd62d7a56ce48187502cd84cab59aa6","source":{"kind":"arxiv","id":"2606.21359","version":1},"attestation_state":"computed","paper":{"title":"Finetuning with Scientific Data Increases Hallucinations: A Multi-domain Factuality Evaluation of LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ekaterina Borisova, Fabio Barth, Georg Rehm, Nikolas Rauscher, Raia Abu Ahmad, Sebastian M\\\"oller","submitted_at":"2026-06-19T12:03:02Z","abstract_excerpt":"Large language models (LLMs) are increasingly used to communicate and explain scientific concepts, yet their tendency to hallucinate poses significant risks in this high stakes use-case. Prior hallucination evaluation work remains largely restricted to the biomedical domain, treats hallucination as a binary task, and has not examined the growing family of scientifically fine-tuned LLMs. We address these gaps with SciFactCheck, a benchmark of 2,500 prompts across five scientific domains, paired with a modular evaluation framework targeting three factuality hallucination types: unverifiability, "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.21359","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-19T12:03:02Z","cross_cats_sorted":[],"title_canon_sha256":"f264f9ac74457061cf0be12122d1f1262ff2c9977d74fe0da7ce98a83c7f4af4","abstract_canon_sha256":"792b74dabc36ed455da5f2359a0089ffa1e9915f83ab505d6bc94c1be6fcd0ff"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:13:07.675826Z","signature_b64":"ydMP/+v4xHRSfaOKfYGAnD3ZG2z7gMfsO1hvolvSHutM7AZ7OU8Z/vCihMmpIx3iWsLz8VnmLzMLpWuarngNAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"69683a40f38245983653c12acf07715a2fd62d7a56ce48187502cd84cab59aa6","last_reissued_at":"2026-06-23T01:13:07.675151Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:13:07.675151Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Finetuning with Scientific Data Increases Hallucinations: A Multi-domain Factuality Evaluation of LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ekaterina Borisova, Fabio Barth, Georg Rehm, Nikolas Rauscher, Raia Abu Ahmad, Sebastian M\\\"oller","submitted_at":"2026-06-19T12:03:02Z","abstract_excerpt":"Large language models (LLMs) are increasingly used to communicate and explain scientific concepts, yet their tendency to hallucinate poses significant risks in this high stakes use-case. Prior hallucination evaluation work remains largely restricted to the biomedical domain, treats hallucination as a binary task, and has not examined the growing family of scientifically fine-tuned LLMs. We address these gaps with SciFactCheck, a benchmark of 2,500 prompts across five scientific domains, paired with a modular evaluation framework targeting three factuality hallucination types: unverifiability, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21359","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.21359/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.21359","created_at":"2026-06-23T01:13:07.675228+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.21359v1","created_at":"2026-06-23T01:13:07.675228+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21359","created_at":"2026-06-23T01:13:07.675228+00:00"},{"alias_kind":"pith_short_12","alias_value":"NFUDUQHTQJCZ","created_at":"2026-06-23T01:13:07.675228+00:00"},{"alias_kind":"pith_short_16","alias_value":"NFUDUQHTQJCZQNST","created_at":"2026-06-23T01:13:07.675228+00:00"},{"alias_kind":"pith_short_8","alias_value":"NFUDUQHT","created_at":"2026-06-23T01:13:07.675228+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI","json":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI.json","graph_json":"https://pith.science/api/pith-number/NFUDUQHTQJCZQNSTYEVM6B3RLI/graph.json","events_json":"https://pith.science/api/pith-number/NFUDUQHTQJCZQNSTYEVM6B3RLI/events.json","paper":"https://pith.science/paper/NFUDUQHT"},"agent_actions":{"view_html":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI","download_json":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI.json","view_paper":"https://pith.science/paper/NFUDUQHT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.21359&json=true","fetch_graph":"https://pith.science/api/pith-number/NFUDUQHTQJCZQNSTYEVM6B3RLI/graph.json","fetch_events":"https://pith.science/api/pith-number/NFUDUQHTQJCZQNSTYEVM6B3RLI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI/action/storage_attestation","attest_author":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI/action/author_attestation","sign_citation":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI/action/citation_signature","submit_replication":"https://pith.science/pith/NFUDUQHTQJCZQNSTYEVM6B3RLI/action/replication_record"}},"created_at":"2026-06-23T01:13:07.675228+00:00","updated_at":"2026-06-23T01:13:07.675228+00:00"}