{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:H5VOCR2KU2AAE6US7JQ5QKBRKG","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":"f54225b6d9870aaabfcc4941a911a6ba75a46d3b741391494bdfdb8553ac8d33","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-31T23:52:06Z","title_canon_sha256":"ecbabbc25616abd6e1b1987993173fef8e0d3dc5cc9320792c5cdc95479e4224"},"schema_version":"1.0","source":{"id":"2411.00264","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.00264","created_at":"2026-07-05T10:47:56Z"},{"alias_kind":"arxiv_version","alias_value":"2411.00264v2","created_at":"2026-07-05T10:47:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.00264","created_at":"2026-07-05T10:47:56Z"},{"alias_kind":"pith_short_12","alias_value":"H5VOCR2KU2AA","created_at":"2026-07-05T10:47:56Z"},{"alias_kind":"pith_short_16","alias_value":"H5VOCR2KU2AAE6US","created_at":"2026-07-05T10:47:56Z"},{"alias_kind":"pith_short_8","alias_value":"H5VOCR2K","created_at":"2026-07-05T10:47:56Z"}],"graph_snapshots":[{"event_id":"sha256:95d12bed8952fdbb8978921a1dacb657e139af8c704e308046f62fead14729e4","target":"graph","created_at":"2026-07-05T10:47:56Z","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/2411.00264/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Humans have the ability to reason about geometric patterns in images and scenes from a young age. However, developing large multimodal models (LMMs) capable of similar reasoning remains a challenge, highlighting the need for robust evaluation methods to assess these capabilities. We introduce \\Turtle, a benchmark designed to evaluate LMMs' capacity to interpret geometric patterns -- given visual examples, textual instructions, or both -- and generate precise code outputs. Inspired by turtle geometry, a notion used to teach children foundational coding and geometric concepts, TurtleBench featur","authors_text":"Sameer Singh, Shayan Doroudi, Sina Rismanchian, Yasaman Razeghi","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-31T23:52:06Z","title":"TurtleBench: A Visual Programming Benchmark in Turtle Geometry"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.00264","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:ed67e534c06724019ac5e5f5bc1be74d228e350e34698e93da67b511b1910a75","target":"record","created_at":"2026-07-05T10:47:56Z","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":"f54225b6d9870aaabfcc4941a911a6ba75a46d3b741391494bdfdb8553ac8d33","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-31T23:52:06Z","title_canon_sha256":"ecbabbc25616abd6e1b1987993173fef8e0d3dc5cc9320792c5cdc95479e4224"},"schema_version":"1.0","source":{"id":"2411.00264","kind":"arxiv","version":2}},"canonical_sha256":"3f6ae1474aa680027a92fa61d82831519b001026098ceabe3815249b1e8cf042","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3f6ae1474aa680027a92fa61d82831519b001026098ceabe3815249b1e8cf042","first_computed_at":"2026-07-05T10:47:56.017311Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:47:56.017311Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hL7+UcZBaVJca1UpamHMQbsa7cPemsWbL+5RVz4eX69vUwkUTQER+1MWzQa9nIJ7qNlNvPd9GzQAXkRTDo8GBg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:47:56.017815Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.00264","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ed67e534c06724019ac5e5f5bc1be74d228e350e34698e93da67b511b1910a75","sha256:95d12bed8952fdbb8978921a1dacb657e139af8c704e308046f62fead14729e4"],"state_sha256":"4ee9229b66fd8851506201bbe4116835d94308e7ae88a68446e6ab984cc567b4"}