{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:46RVYHKLJLC4P7JNJ3Y2OISA7N","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":"0d8196d4e2194a10d200a00731a11458fbec3120ba598e7b1bb8c4c61ff1521e","cross_cats_sorted":["cs.AI","cs.CV","cs.HC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-07-25T23:57:44Z","title_canon_sha256":"b1e98ec9936ae0a928ef9cb7f98ac7f498c7ee567c02eefe11b5aecac6ddbdf0"},"schema_version":"1.0","source":{"id":"2207.12576","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2207.12576","created_at":"2026-07-05T05:05:08Z"},{"alias_kind":"arxiv_version","alias_value":"2207.12576v2","created_at":"2026-07-05T05:05:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.12576","created_at":"2026-07-05T05:05:08Z"},{"alias_kind":"pith_short_12","alias_value":"46RVYHKLJLC4","created_at":"2026-07-05T05:05:08Z"},{"alias_kind":"pith_short_16","alias_value":"46RVYHKLJLC4P7JN","created_at":"2026-07-05T05:05:08Z"},{"alias_kind":"pith_short_8","alias_value":"46RVYHKL","created_at":"2026-07-05T05:05:08Z"}],"graph_snapshots":[{"event_id":"sha256:c5dc26a50a2700ad1c33d5277a3aab4906ff2807068f1d8daae98bf2c378c618","target":"graph","created_at":"2026-07-05T05:05:08Z","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/2207.12576/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While vision-and-language models perform well on tasks such as visual question answering, they struggle when it comes to basic human commonsense reasoning skills. In this work, we introduce WinoGAViL: an online game of vision-and-language associations (e.g., between werewolves and a full moon), used as a dynamic evaluation benchmark. Inspired by the popular card game Codenames, a spymaster gives a textual cue related to several visual candidates, and another player tries to identify them. Human players are rewarded for creating associations that are challenging for a rival AI model but still s","authors_text":"Gabriel Stanovsky, Mohit Bansal, Nitzan Bitton Guetta, Ron Yosef, Roy Schwartz, Yonatan Bitton, Yuval Elovici","cross_cats":["cs.AI","cs.CV","cs.HC"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-07-25T23:57:44Z","title":"WinoGAViL: Gamified Association Benchmark to Challenge Vision-and-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.12576","kind":"arxiv","version":2},"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":"sha256:0a4997e5fc94e8e0d9224d7d6624833d8dc3af03656c927a4fdc5e7e7e1b349d","target":"record","created_at":"2026-07-05T05:05:08Z","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":"0d8196d4e2194a10d200a00731a11458fbec3120ba598e7b1bb8c4c61ff1521e","cross_cats_sorted":["cs.AI","cs.CV","cs.HC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-07-25T23:57:44Z","title_canon_sha256":"b1e98ec9936ae0a928ef9cb7f98ac7f498c7ee567c02eefe11b5aecac6ddbdf0"},"schema_version":"1.0","source":{"id":"2207.12576","kind":"arxiv","version":2}},"canonical_sha256":"e7a35c1d4b4ac5c7fd2d4ef1a72240fb5f8b9350b238b42a415d787c8ae513d3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e7a35c1d4b4ac5c7fd2d4ef1a72240fb5f8b9350b238b42a415d787c8ae513d3","first_computed_at":"2026-07-05T05:05:08.024946Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:05:08.024946Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1DOeIuSQPjWU7FejsvQ/kdaLmtlPvymPBUd3lmUnfrr6go6o6E8Qfd/AzWQ93TL6v9mCI2v+kLd35jEyyMldBw==","signature_status":"signed_v1","signed_at":"2026-07-05T05:05:08.025402Z","signed_message":"canonical_sha256_bytes"},"source_id":"2207.12576","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0a4997e5fc94e8e0d9224d7d6624833d8dc3af03656c927a4fdc5e7e7e1b349d","sha256:c5dc26a50a2700ad1c33d5277a3aab4906ff2807068f1d8daae98bf2c378c618"],"state_sha256":"ea29582411bf6e911af013f847234a9cb1aea1b90f1a0e7c225df0985da0d84c"}