{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:ZMZBLCC4S3ADLZZFFFDFK4AKWG","short_pith_number":"pith:ZMZBLCC4","canonical_record":{"source":{"id":"2203.07613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-15T03:01:03Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"e70845545d3e469f764f5b3a25bcc41f5f1d15462ab87d7157932218146b1700","abstract_canon_sha256":"a3370630fb2bf1b5ce6c832dfa3cc09e589fa1adb81cc2e88dc20b3f5a3cff16"},"schema_version":"1.0"},"canonical_sha256":"cb3215885c96c035e725294655700ab18f27fb1186bb4d8e025a3d07bf28bca1","source":{"kind":"arxiv","id":"2203.07613","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.07613","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"arxiv_version","alias_value":"2203.07613v1","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.07613","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"pith_short_12","alias_value":"ZMZBLCC4S3AD","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"pith_short_16","alias_value":"ZMZBLCC4S3ADLZZF","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"pith_short_8","alias_value":"ZMZBLCC4","created_at":"2026-07-05T04:05:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:ZMZBLCC4S3ADLZZFFFDFK4AKWG","target":"record","payload":{"canonical_record":{"source":{"id":"2203.07613","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-15T03:01:03Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"e70845545d3e469f764f5b3a25bcc41f5f1d15462ab87d7157932218146b1700","abstract_canon_sha256":"a3370630fb2bf1b5ce6c832dfa3cc09e589fa1adb81cc2e88dc20b3f5a3cff16"},"schema_version":"1.0"},"canonical_sha256":"cb3215885c96c035e725294655700ab18f27fb1186bb4d8e025a3d07bf28bca1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:05:20.550906Z","signature_b64":"4gto0fySXYaENJffLuVvcz4mNdIUwr9mQNVPVAXsOjcSsr2jrB4aD+XeIBi/Ic4iaQe2ZKnwIu2tD+cV8c8ODA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cb3215885c96c035e725294655700ab18f27fb1186bb4d8e025a3d07bf28bca1","last_reissued_at":"2026-07-05T04:05:20.550527Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:05:20.550527Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2203.07613","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":"2026-07-05T04:05:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Gwla1LUgvWLTiePDwxD27xTAzZJiPqvDUsjux+qgebsOSMSoPrIyAsYihYzHZVzoaJR/z6QCZRI+AEMUYZPyAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T09:32:12.747565Z"},"content_sha256":"bbd430f131edb703e87fb1b809d9dd49df52dd2a363c6fbff91e9db113bd094a","schema_version":"1.0","event_id":"sha256:bbd430f131edb703e87fb1b809d9dd49df52dd2a363c6fbff91e9db113bd094a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:ZMZBLCC4S3ADLZZFFFDFK4AKWG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CARETS: A Consistency And Robustness Evaluative Test Suite for VQA","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Carlos E. Jimenez, Karthik Narasimhan, Olga Russakovsky","submitted_at":"2022-03-15T03:01:03Z","abstract_excerpt":"We introduce CARETS, a systematic test suite to measure consistency and robustness of modern VQA models through a series of six fine-grained capability tests. In contrast to existing VQA test sets, CARETS features balanced question generation to create pairs of instances to test models, with each pair focusing on a specific capability such as rephrasing, logical symmetry or image obfuscation. We evaluate six modern VQA systems on CARETS and identify several actionable weaknesses in model comprehension, especially with concepts such as negation, disjunction, or hypernym invariance. Interestingl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.07613","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/2203.07613/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":"2026-07-05T04:05:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L8gzyV7bstC4uAv7ZYKxr9TLs2QudblMWaZvVilvL0PGMFHXD+I+vjXI53lhc1QPp9ZLHX0d+cEVzXZf5T3oBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T09:32:12.748062Z"},"content_sha256":"ca09beddf25baf9da3794669f81626b17c23e07eff2d0a78a674ce532cf61207","schema_version":"1.0","event_id":"sha256:ca09beddf25baf9da3794669f81626b17c23e07eff2d0a78a674ce532cf61207"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZMZBLCC4S3ADLZZFFFDFK4AKWG/bundle.json","state_url":"https://pith.science/pith/ZMZBLCC4S3ADLZZFFFDFK4AKWG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZMZBLCC4S3ADLZZFFFDFK4AKWG/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":"2026-08-21T09:32:12Z","links":{"resolver":"https://pith.science/pith/ZMZBLCC4S3ADLZZFFFDFK4AKWG","bundle":"https://pith.science/pith/ZMZBLCC4S3ADLZZFFFDFK4AKWG/bundle.json","state":"https://pith.science/pith/ZMZBLCC4S3ADLZZFFFDFK4AKWG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZMZBLCC4S3ADLZZFFFDFK4AKWG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:ZMZBLCC4S3ADLZZFFFDFK4AKWG","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":"a3370630fb2bf1b5ce6c832dfa3cc09e589fa1adb81cc2e88dc20b3f5a3cff16","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-15T03:01:03Z","title_canon_sha256":"e70845545d3e469f764f5b3a25bcc41f5f1d15462ab87d7157932218146b1700"},"schema_version":"1.0","source":{"id":"2203.07613","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2203.07613","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"arxiv_version","alias_value":"2203.07613v1","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.07613","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"pith_short_12","alias_value":"ZMZBLCC4S3AD","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"pith_short_16","alias_value":"ZMZBLCC4S3ADLZZF","created_at":"2026-07-05T04:05:20Z"},{"alias_kind":"pith_short_8","alias_value":"ZMZBLCC4","created_at":"2026-07-05T04:05:20Z"}],"graph_snapshots":[{"event_id":"sha256:ca09beddf25baf9da3794669f81626b17c23e07eff2d0a78a674ce532cf61207","target":"graph","created_at":"2026-07-05T04:05:20Z","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/2203.07613/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce CARETS, a systematic test suite to measure consistency and robustness of modern VQA models through a series of six fine-grained capability tests. In contrast to existing VQA test sets, CARETS features balanced question generation to create pairs of instances to test models, with each pair focusing on a specific capability such as rephrasing, logical symmetry or image obfuscation. We evaluate six modern VQA systems on CARETS and identify several actionable weaknesses in model comprehension, especially with concepts such as negation, disjunction, or hypernym invariance. Interestingl","authors_text":"Carlos E. Jimenez, Karthik Narasimhan, Olga Russakovsky","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-15T03:01:03Z","title":"CARETS: A Consistency And Robustness Evaluative Test Suite for VQA"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.07613","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:bbd430f131edb703e87fb1b809d9dd49df52dd2a363c6fbff91e9db113bd094a","target":"record","created_at":"2026-07-05T04:05:20Z","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":"a3370630fb2bf1b5ce6c832dfa3cc09e589fa1adb81cc2e88dc20b3f5a3cff16","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-03-15T03:01:03Z","title_canon_sha256":"e70845545d3e469f764f5b3a25bcc41f5f1d15462ab87d7157932218146b1700"},"schema_version":"1.0","source":{"id":"2203.07613","kind":"arxiv","version":1}},"canonical_sha256":"cb3215885c96c035e725294655700ab18f27fb1186bb4d8e025a3d07bf28bca1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cb3215885c96c035e725294655700ab18f27fb1186bb4d8e025a3d07bf28bca1","first_computed_at":"2026-07-05T04:05:20.550527Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:05:20.550527Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4gto0fySXYaENJffLuVvcz4mNdIUwr9mQNVPVAXsOjcSsr2jrB4aD+XeIBi/Ic4iaQe2ZKnwIu2tD+cV8c8ODA==","signature_status":"signed_v1","signed_at":"2026-07-05T04:05:20.550906Z","signed_message":"canonical_sha256_bytes"},"source_id":"2203.07613","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bbd430f131edb703e87fb1b809d9dd49df52dd2a363c6fbff91e9db113bd094a","sha256:ca09beddf25baf9da3794669f81626b17c23e07eff2d0a78a674ce532cf61207"],"state_sha256":"818bca718d384c9fe250c3e1e30fb73a1dd88c0b88eb6eba55bf985678a5411f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W0RQNGnkvfdGPLcEvWsrjEOkMG1eCP7+gAMGWd4KSiv/wmPWxEWBUVD8Ll+imTdSMjtH2RYZnxjJKvgvm6idDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T09:32:12.751961Z","bundle_sha256":"3deb82b9f55b396a5b080d257f4b84350191042371e7684cd409eb74842ce9b7"}}