{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HUXODFEQ7DRBRSHMAFWKIR5MBX","short_pith_number":"pith:HUXODFEQ","schema_version":"1.0","canonical_sha256":"3d2ee19490f8e218c8ec016ca447ac0df9deac83848fa09996f8367bea1ee8ff","source":{"kind":"arxiv","id":"2606.24115","version":1},"attestation_state":"computed","paper":{"title":"A Benchmark for Hallucination Detection in VLMs for Gastrointestinal Endoscopy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Aminu Lawal, Binod Bhattarai, Maria Carmen Romano, Niyoj Oli, Prashnna Gyawali, Sachin Acharya","submitted_at":"2026-06-23T04:04:19Z","abstract_excerpt":"Vision-language models (VLMs) are prone to hallucination, which remains a major barrier to their safe deployment in clinical practice. To date, most hallucination detection methods have been evaluated on radiology benchmarks such as MIMIC-CXR and VQA-RAD, while gastrointestinal (GI) endoscopy remains largely underexplored. In this paper, we benchmark nine hallucination detection methods on the Gut-VLM dataset, a GI diagnostic Visual Question Answering (VQA) dataset with 4,392 test VQA pairs, across five VLMs (MedGemma-4B, MedGemma-27B, LLaVA-Med-7B, LLaVA-v1.6-7B, and Lingshu-32B). The methods"},"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":"2606.24115","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-23T04:04:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cb2a0b6edd5df484799917c112e80042d516dee46b76dd153f7f0840bf24ec5b","abstract_canon_sha256":"3c6a0bb1a96fe1fbdd41e3e097e60178e1bc31304b887fb2f00e10d15e9611e1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T01:14:41.499583Z","signature_b64":"JbUQZF9Rdj50RUWUWCyQSmFzX9Edh+1dVxXN7+bRGEqytziZNdyz6ZIPw5GXsGhZsJdz+ck3IYouWpc9T7ovBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d2ee19490f8e218c8ec016ca447ac0df9deac83848fa09996f8367bea1ee8ff","last_reissued_at":"2026-06-24T01:14:41.499202Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T01:14:41.499202Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Benchmark for Hallucination Detection in VLMs for Gastrointestinal Endoscopy","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Aminu Lawal, Binod Bhattarai, Maria Carmen Romano, Niyoj Oli, Prashnna Gyawali, Sachin Acharya","submitted_at":"2026-06-23T04:04:19Z","abstract_excerpt":"Vision-language models (VLMs) are prone to hallucination, which remains a major barrier to their safe deployment in clinical practice. To date, most hallucination detection methods have been evaluated on radiology benchmarks such as MIMIC-CXR and VQA-RAD, while gastrointestinal (GI) endoscopy remains largely underexplored. In this paper, we benchmark nine hallucination detection methods on the Gut-VLM dataset, a GI diagnostic Visual Question Answering (VQA) dataset with 4,392 test VQA pairs, across five VLMs (MedGemma-4B, MedGemma-27B, LLaVA-Med-7B, LLaVA-v1.6-7B, and Lingshu-32B). The methods"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24115","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/2606.24115/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":"2606.24115","created_at":"2026-06-24T01:14:41.499266+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.24115v1","created_at":"2026-06-24T01:14:41.499266+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24115","created_at":"2026-06-24T01:14:41.499266+00:00"},{"alias_kind":"pith_short_12","alias_value":"HUXODFEQ7DRB","created_at":"2026-06-24T01:14:41.499266+00:00"},{"alias_kind":"pith_short_16","alias_value":"HUXODFEQ7DRBRSHM","created_at":"2026-06-24T01:14:41.499266+00:00"},{"alias_kind":"pith_short_8","alias_value":"HUXODFEQ","created_at":"2026-06-24T01:14:41.499266+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/HUXODFEQ7DRBRSHMAFWKIR5MBX","json":"https://pith.science/pith/HUXODFEQ7DRBRSHMAFWKIR5MBX.json","graph_json":"https://pith.science/api/pith-number/HUXODFEQ7DRBRSHMAFWKIR5MBX/graph.json","events_json":"https://pith.science/api/pith-number/HUXODFEQ7DRBRSHMAFWKIR5MBX/events.json","paper":"https://pith.science/paper/HUXODFEQ"},"agent_actions":{"view_html":"https://pith.science/pith/HUXODFEQ7DRBRSHMAFWKIR5MBX","download_json":"https://pith.science/pith/HUXODFEQ7DRBRSHMAFWKIR5MBX.json","view_paper":"https://pith.science/paper/HUXODFEQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.24115&json=true","fetch_graph":"https://pith.science/api/pith-number/HUXODFEQ7DRBRSHMAFWKIR5MBX/graph.json","fetch_events":"https://pith.science/api/pith-number/HUXODFEQ7DRBRSHMAFWKIR5MBX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HUXODFEQ7DRBRSHMAFWKIR5MBX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HUXODFEQ7DRBRSHMAFWKIR5MBX/action/storage_attestation","attest_author":"https://pith.science/pith/HUXODFEQ7DRBRSHMAFWKIR5MBX/action/author_attestation","sign_citation":"https://pith.science/pith/HUXODFEQ7DRBRSHMAFWKIR5MBX/action/citation_signature","submit_replication":"https://pith.science/pith/HUXODFEQ7DRBRSHMAFWKIR5MBX/action/replication_record"}},"created_at":"2026-06-24T01:14:41.499266+00:00","updated_at":"2026-06-24T01:14:41.499266+00:00"}