{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6HA2OSW5XCD45PDD4JY3R33HML","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"b26d7d984fdceb8d1673125284ee658da0cbfad6fbd61158adb80d0e4ddca9ed","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-26T07:02:22Z","title_canon_sha256":"f15aaa9b24d2619f7ac9a77f4fe279fcd12f67fe63449347ce973885a0c5151b"},"schema_version":"1.0","source":{"id":"2509.21979","kind":"arxiv","version":6}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.21979","created_at":"2026-05-20T00:04:13Z"},{"alias_kind":"arxiv_version","alias_value":"2509.21979v6","created_at":"2026-05-20T00:04:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.21979","created_at":"2026-05-20T00:04:13Z"},{"alias_kind":"pith_short_12","alias_value":"6HA2OSW5XCD4","created_at":"2026-05-20T00:04:13Z"},{"alias_kind":"pith_short_16","alias_value":"6HA2OSW5XCD45PDD","created_at":"2026-05-20T00:04:13Z"},{"alias_kind":"pith_short_8","alias_value":"6HA2OSW5","created_at":"2026-05-20T00:04:13Z"}],"graph_snapshots":[{"event_id":"sha256:85052ab1003cf043e80657ddc5c1588eeceada44d166810f8c6d444e0feb7e80","target":"graph","created_at":"2026-05-20T00:04:13Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Current VLMs are highly susceptible to visual cues, with failure rates showing a correlation to model size or overall accuracy; perceived authority and user mimicry are powerful triggers suggesting a bias mechanism independent of visual data; VIPER reduces sycophancy while maintaining interpretability and consistently outperforms baseline methods."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The hierarchical medical visual question answering templates and authority/mimicry triggers accurately capture real-world sycophancy without introducing artificial biases that would not appear in actual clinical interactions (stated in the abstract description of the benchmark construction)."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Introduces a medical sycophancy benchmark for VLMs and the VIPER strategy to reduce agreement with non-evidence cues while preserving interpretability."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Medical vision language models exhibit sycophancy driven by visual cues and authority signals, which a filtering strategy called VIPER can reduce."}],"snapshot_sha256":"a5caaf94286556142a2ac39956165aef603d47cb41dd5eb1ce9a7cb300fde720"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"e67f19ec4f8d73d626a689b4f1ed1f43ccb77231bd1b3d5840f8aa1ddabb51f1"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2509.21979/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Visual language models (VLMs) have the potential to transform medical workflows. However, the deployment is limited by sycophancy. Despite this serious threat to patient safety, a systematic benchmark remains lacking. This paper addresses this gap by introducing a Medical benchmark that applies multiple templates to VLMs in a hierarchical medical visual question answering task. We find that current VLMs are highly susceptible to visual cues, with failure rates showing a correlation to model size or overall accuracy. we discover that perceived authority and user mimicry are powerful triggers, s","authors_text":"Di Wang, Hongbin Lin, Jingwei Lv, Juangui Xu, Jun Wen, Lijie Hu, Shu Yang, Xinyue Xu, Zikun Guo","cross_cats":["cs.AI"],"headline":"Medical vision language models exhibit sycophancy driven by visual cues and authority signals, which a filtering strategy called VIPER can reduce.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-26T07:02:22Z","title":"Benchmarking and Mitigating Sycophancy in Medical Vision Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.21979","kind":"arxiv","version":6},"verdict":{"created_at":"2026-05-18T13:58:53.537348Z","id":"e82a390c-8ed6-4f54-9a93-c240b871bc79","model_set":{"reader":"grok-4.3"},"one_line_summary":"Introduces a medical sycophancy benchmark for VLMs and the VIPER strategy to reduce agreement with non-evidence cues while preserving interpretability.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Medical vision language models exhibit sycophancy driven by visual cues and authority signals, which a filtering strategy called VIPER can reduce.","strongest_claim":"Current VLMs are highly susceptible to visual cues, with failure rates showing a correlation to model size or overall accuracy; perceived authority and user mimicry are powerful triggers suggesting a bias mechanism independent of visual data; VIPER reduces sycophancy while maintaining interpretability and consistently outperforms baseline methods.","weakest_assumption":"The hierarchical medical visual question answering templates and authority/mimicry triggers accurately capture real-world sycophancy without introducing artificial biases that would not appear in actual clinical interactions (stated in the abstract description of the benchmark construction)."}},"verdict_id":"e82a390c-8ed6-4f54-9a93-c240b871bc79"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:fc7613e4b3d4f935b17295f634ec3f4cc6888382acb903453f2c1b526d07ac23","target":"record","created_at":"2026-05-20T00:04:13Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"b26d7d984fdceb8d1673125284ee658da0cbfad6fbd61158adb80d0e4ddca9ed","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-09-26T07:02:22Z","title_canon_sha256":"f15aaa9b24d2619f7ac9a77f4fe279fcd12f67fe63449347ce973885a0c5151b"},"schema_version":"1.0","source":{"id":"2509.21979","kind":"arxiv","version":6}},"canonical_sha256":"f1c1a74addb887cebc63e271b8ef6762d0c3484f5d7f5da53b6d42b4b24fa05d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f1c1a74addb887cebc63e271b8ef6762d0c3484f5d7f5da53b6d42b4b24fa05d","first_computed_at":"2026-05-20T00:04:13.908972Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:13.908972Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VaGj+zl0Glqq3l71AU63zERXXWgo9juRi1YhA9UiwnVhsLBbR+77vzRe/Gvy1iXucmDhI1Qqgk/+toxdRx3yAw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:13.909747Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.21979","source_kind":"arxiv","source_version":6}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fc7613e4b3d4f935b17295f634ec3f4cc6888382acb903453f2c1b526d07ac23","sha256:85052ab1003cf043e80657ddc5c1588eeceada44d166810f8c6d444e0feb7e80"],"state_sha256":"bed232483aeaac291b90c17d6b3ecaa699a1e5d8e62b69a01219b1d4ffe9798e"}