{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TXE6TRFJRNG2KJ2UKNLADZV7T5","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":"762c98df81dc101dc55fb51622b7c1d37538b09a93fbf4290cd92ec3c9fcf6dc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-09T20:53:43Z","title_canon_sha256":"2fc4fc97b8373a5b3ee53e318d097f01e348dbef4cf1b5294f298a4292f94cfe"},"schema_version":"1.0","source":{"id":"2506.08227","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.08227","created_at":"2026-07-05T11:18:46Z"},{"alias_kind":"arxiv_version","alias_value":"2506.08227v1","created_at":"2026-07-05T11:18:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.08227","created_at":"2026-07-05T11:18:46Z"},{"alias_kind":"pith_short_12","alias_value":"TXE6TRFJRNG2","created_at":"2026-07-05T11:18:46Z"},{"alias_kind":"pith_short_16","alias_value":"TXE6TRFJRNG2KJ2U","created_at":"2026-07-05T11:18:46Z"},{"alias_kind":"pith_short_8","alias_value":"TXE6TRFJ","created_at":"2026-07-05T11:18:46Z"}],"graph_snapshots":[{"event_id":"sha256:ada44de40260456fb9d5e40aeef6dbd4c71c530e323df1e42c4b6faf19ff5d7d","target":"graph","created_at":"2026-07-05T11:18:46Z","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/2506.08227/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We investigate 17 benchmarks (e.g. SugarCREPE, VALSE) commonly used for measuring compositional understanding capabilities of vision-language models (VLMs). We scrutinize design choices in their construction, including data source (e.g. MS-COCO) and curation procedures (e.g. constructing negative images/captions), uncovering several inherent biases across most benchmarks. We find that blind heuristics (e.g. token-length, log-likelihood under a language model) perform on par with CLIP models, indicating that these benchmarks do not effectively measure compositional understanding. We demonstrate","authors_text":"Jenia Jitsev, Matthias Bethge, Mehdi Cherti, Samuel Albanie, Shyamgopal Karthik, Vishaal Udandarao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-09T20:53:43Z","title":"A Good CREPE needs more than just Sugar: Investigating Biases in Compositional Vision-Language Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.08227","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":"sha256:204be5a4eb188a3777244e3ac41d9d7c9f717af79856c6052432d52d1f5a787f","target":"record","created_at":"2026-07-05T11:18:46Z","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":"762c98df81dc101dc55fb51622b7c1d37538b09a93fbf4290cd92ec3c9fcf6dc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-09T20:53:43Z","title_canon_sha256":"2fc4fc97b8373a5b3ee53e318d097f01e348dbef4cf1b5294f298a4292f94cfe"},"schema_version":"1.0","source":{"id":"2506.08227","kind":"arxiv","version":1}},"canonical_sha256":"9dc9e9c4a98b4da52754535601e6bf9f5075e9754232800fe3384b53861b1835","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9dc9e9c4a98b4da52754535601e6bf9f5075e9754232800fe3384b53861b1835","first_computed_at":"2026-07-05T11:18:46.507534Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:18:46.507534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Qbfh5Q1KDAJ3rl+IgJajKwuipgXhtPoncxImSW69EsmcWNAgDamJc+nfWxVo/pvd/VnPx7M9A+2oVZjiEpOlAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:18:46.507991Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.08227","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:204be5a4eb188a3777244e3ac41d9d7c9f717af79856c6052432d52d1f5a787f","sha256:ada44de40260456fb9d5e40aeef6dbd4c71c530e323df1e42c4b6faf19ff5d7d"],"state_sha256":"a1a027aecde1927d5ae8ff84ff1c01362156b93a0404a210c476a206e5abeca1"}