{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:J56HUPFNEX7KN6TZ5LSJBMGXXI","short_pith_number":"pith:J56HUPFN","canonical_record":{"source":{"id":"2311.16101","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-11-27T18:59:42Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"8b187a404d2c16c500b4932e7df42905c6f51bc0c96c90f131b6af2eca14cfd8","abstract_canon_sha256":"44c2d8ada9f17b9a37fa668cf645b3c59cf14c42399de4ea2d24a30410699b92"},"schema_version":"1.0"},"canonical_sha256":"4f7c7a3cad25fea6fa79eae490b0d7ba2518dc2036918384729667506413a46d","source":{"kind":"arxiv","id":"2311.16101","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.16101","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"arxiv_version","alias_value":"2311.16101v1","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.16101","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"pith_short_12","alias_value":"J56HUPFNEX7K","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"pith_short_16","alias_value":"J56HUPFNEX7KN6TZ","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"pith_short_8","alias_value":"J56HUPFN","created_at":"2026-07-05T07:17:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:J56HUPFNEX7KN6TZ5LSJBMGXXI","target":"record","payload":{"canonical_record":{"source":{"id":"2311.16101","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-11-27T18:59:42Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"8b187a404d2c16c500b4932e7df42905c6f51bc0c96c90f131b6af2eca14cfd8","abstract_canon_sha256":"44c2d8ada9f17b9a37fa668cf645b3c59cf14c42399de4ea2d24a30410699b92"},"schema_version":"1.0"},"canonical_sha256":"4f7c7a3cad25fea6fa79eae490b0d7ba2518dc2036918384729667506413a46d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:17:20.390734Z","signature_b64":"XDciBX70xQ3XR7hIJ/NSPGiZXR1sjTEqCcF9IGpNSfqhQ2eZSlKNbY9l32txuZgKgvZv/WiJ/vOLw9ma3BlUDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4f7c7a3cad25fea6fa79eae490b0d7ba2518dc2036918384729667506413a46d","last_reissued_at":"2026-07-05T07:17:20.390214Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:17:20.390214Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2311.16101","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:17:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gUJx6hPdoupmmsuvhNmCXUbr6/bpfevesA7TCrkbjRveUYlP+Um+zKmPj+7ALLFrChEw8XVS6PEXIP96S0CtBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T22:37:10.346101Z"},"content_sha256":"3531f2e39f910088df45766adb197457f56b9b436f9f26a66f46f45f739d65e3","schema_version":"1.0","event_id":"sha256:3531f2e39f910088df45766adb197457f56b9b436f9f26a66f46f45f739d65e3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:J56HUPFNEX7KN6TZ5LSJBMGXXI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bingchen Zhao, Chenhang Cui, Cihang Xie, Haoqin Tu, Huaxiu Yao, Junlin Han, Wangchunshu Zhou, Yiyang Zhou, Zijun Wang","submitted_at":"2023-11-27T18:59:42Z","abstract_excerpt":"This work focuses on the potential of Vision LLMs (VLLMs) in visual reasoning. Different from prior studies, we shift our focus from evaluating standard performance to introducing a comprehensive safety evaluation suite, covering both out-of-distribution (OOD) generalization and adversarial robustness. For the OOD evaluation, we present two novel VQA datasets, each with one variant, designed to test model performance under challenging conditions. In exploring adversarial robustness, we propose a straightforward attack strategy for misleading VLLMs to produce visual-unrelated responses. Moreove"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.16101","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/2311.16101/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:17:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GOUiF3GcCbQEb7l8gDsR3JqjDJuHXVi+OSsN2t67JmH68TZmJ38vutPCYM/ViJxQDrmmZGdeCebSzxuBnGGSDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T22:37:10.347033Z"},"content_sha256":"219eeccabc44ac019b86fa9ecd664b5f0812fa593e660e4500b2eea3b6634f5f","schema_version":"1.0","event_id":"sha256:219eeccabc44ac019b86fa9ecd664b5f0812fa593e660e4500b2eea3b6634f5f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/J56HUPFNEX7KN6TZ5LSJBMGXXI/bundle.json","state_url":"https://pith.science/pith/J56HUPFNEX7KN6TZ5LSJBMGXXI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/J56HUPFNEX7KN6TZ5LSJBMGXXI/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T22:37:10Z","links":{"resolver":"https://pith.science/pith/J56HUPFNEX7KN6TZ5LSJBMGXXI","bundle":"https://pith.science/pith/J56HUPFNEX7KN6TZ5LSJBMGXXI/bundle.json","state":"https://pith.science/pith/J56HUPFNEX7KN6TZ5LSJBMGXXI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/J56HUPFNEX7KN6TZ5LSJBMGXXI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:J56HUPFNEX7KN6TZ5LSJBMGXXI","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":"44c2d8ada9f17b9a37fa668cf645b3c59cf14c42399de4ea2d24a30410699b92","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-11-27T18:59:42Z","title_canon_sha256":"8b187a404d2c16c500b4932e7df42905c6f51bc0c96c90f131b6af2eca14cfd8"},"schema_version":"1.0","source":{"id":"2311.16101","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.16101","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"arxiv_version","alias_value":"2311.16101v1","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.16101","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"pith_short_12","alias_value":"J56HUPFNEX7K","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"pith_short_16","alias_value":"J56HUPFNEX7KN6TZ","created_at":"2026-07-05T07:17:20Z"},{"alias_kind":"pith_short_8","alias_value":"J56HUPFN","created_at":"2026-07-05T07:17:20Z"}],"graph_snapshots":[{"event_id":"sha256:219eeccabc44ac019b86fa9ecd664b5f0812fa593e660e4500b2eea3b6634f5f","target":"graph","created_at":"2026-07-05T07:17:20Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2311.16101/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This work focuses on the potential of Vision LLMs (VLLMs) in visual reasoning. Different from prior studies, we shift our focus from evaluating standard performance to introducing a comprehensive safety evaluation suite, covering both out-of-distribution (OOD) generalization and adversarial robustness. For the OOD evaluation, we present two novel VQA datasets, each with one variant, designed to test model performance under challenging conditions. In exploring adversarial robustness, we propose a straightforward attack strategy for misleading VLLMs to produce visual-unrelated responses. Moreove","authors_text":"Bingchen Zhao, Chenhang Cui, Cihang Xie, Haoqin Tu, Huaxiu Yao, Junlin Han, Wangchunshu Zhou, Yiyang Zhou, Zijun Wang","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-11-27T18:59:42Z","title":"How Many Unicorns Are in This Image? A Safety Evaluation Benchmark for Vision LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.16101","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3531f2e39f910088df45766adb197457f56b9b436f9f26a66f46f45f739d65e3","target":"record","created_at":"2026-07-05T07:17:20Z","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":"44c2d8ada9f17b9a37fa668cf645b3c59cf14c42399de4ea2d24a30410699b92","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-11-27T18:59:42Z","title_canon_sha256":"8b187a404d2c16c500b4932e7df42905c6f51bc0c96c90f131b6af2eca14cfd8"},"schema_version":"1.0","source":{"id":"2311.16101","kind":"arxiv","version":1}},"canonical_sha256":"4f7c7a3cad25fea6fa79eae490b0d7ba2518dc2036918384729667506413a46d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4f7c7a3cad25fea6fa79eae490b0d7ba2518dc2036918384729667506413a46d","first_computed_at":"2026-07-05T07:17:20.390214Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:17:20.390214Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XDciBX70xQ3XR7hIJ/NSPGiZXR1sjTEqCcF9IGpNSfqhQ2eZSlKNbY9l32txuZgKgvZv/WiJ/vOLw9ma3BlUDg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:17:20.390734Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.16101","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3531f2e39f910088df45766adb197457f56b9b436f9f26a66f46f45f739d65e3","sha256:219eeccabc44ac019b86fa9ecd664b5f0812fa593e660e4500b2eea3b6634f5f"],"state_sha256":"104875907f3a67d447088ae63c7281760e2653dc4f13223ab95163450b656078"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gP9STY7KP1uw5OVaMAcXqITmB7tlZntFHrMnYIG5LYS3SmcO1GJY0I9vdvYngUI/2un6L3t6d4m+2khkGWrUBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T22:37:10.354456Z","bundle_sha256":"3f511bf08b4cbdfd8b20100fa2a393cbc7afdae6adc1fe09b07913d5cf2c1745"}}