{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:Q7R4TJMRXURKWJ7OQQHR4R7VJO","short_pith_number":"pith:Q7R4TJMR","canonical_record":{"source":{"id":"2411.18932","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-28T05:51:45Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"c91f09613a68ebabdbe39ace951b3f088a36ff7a2a0dc7e1673ee4a18407b2e9","abstract_canon_sha256":"2a0fcd2a430b310d2b5e7995b0a6157b5cad65900ebe416d08cbc62691ad0169"},"schema_version":"1.0"},"canonical_sha256":"87e3c9a591bd22ab27ee840f1e47f54bbe9ea705f6e050781dc2fd678c3677f4","source":{"kind":"arxiv","id":"2411.18932","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.18932","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"arxiv_version","alias_value":"2411.18932v1","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.18932","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"pith_short_12","alias_value":"Q7R4TJMRXURK","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"pith_short_16","alias_value":"Q7R4TJMRXURKWJ7O","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"pith_short_8","alias_value":"Q7R4TJMR","created_at":"2026-07-05T09:41:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:Q7R4TJMRXURKWJ7OQQHR4R7VJO","target":"record","payload":{"canonical_record":{"source":{"id":"2411.18932","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-28T05:51:45Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"c91f09613a68ebabdbe39ace951b3f088a36ff7a2a0dc7e1673ee4a18407b2e9","abstract_canon_sha256":"2a0fcd2a430b310d2b5e7995b0a6157b5cad65900ebe416d08cbc62691ad0169"},"schema_version":"1.0"},"canonical_sha256":"87e3c9a591bd22ab27ee840f1e47f54bbe9ea705f6e050781dc2fd678c3677f4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:41:54.147408Z","signature_b64":"A6KEkDTpUbd4Hie0qhCJbhHsgzcE0AbcBvZdflo2V/LQBM61x6EQhmK8JDkN9EkU8El2HBpxf9AywbqXUpk4Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87e3c9a591bd22ab27ee840f1e47f54bbe9ea705f6e050781dc2fd678c3677f4","last_reissued_at":"2026-07-05T09:41:54.146944Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:41:54.146944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.18932","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":"2026-07-05T09:41:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p0943JaAvSp6CbZnFoarRCCkGM91sSagfPt55V2ZojbWBNJldeDDOgZAbcRFCm042jeusTr/7O9sWckoNQ97BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T20:20:22.461705Z"},"content_sha256":"d9a26e63efcb6c91184cd96646607fa2836e1fc20156cc1073928bad49f44a1b","schema_version":"1.0","event_id":"sha256:d9a26e63efcb6c91184cd96646607fa2836e1fc20156cc1073928bad49f44a1b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:Q7R4TJMRXURKWJ7OQQHR4R7VJO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ScratchEval: Are GPT-4o Smarter than My Child? Evaluating Large Multimodal Models with Visual Programming Challenges","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Hongzhan Lin, Jing Ma, Rao Fu, Zhen Ye, Ziyang Luo","submitted_at":"2024-11-28T05:51:45Z","abstract_excerpt":"Recent advancements in large multimodal models (LMMs) have showcased impressive code generation capabilities, primarily evaluated through image-to-code benchmarks. However, these benchmarks are limited to specific visual programming scenarios where the logic reasoning and the multimodal understanding capacities are split apart. To fill this gap, we propose ScratchEval, a novel benchmark designed to evaluate the visual programming reasoning ability of LMMs. ScratchEval is based on Scratch, a block-based visual programming language widely used in children's programming education. By integrating "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.18932","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/2411.18932/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":"2026-07-05T09:41:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NEjn7OGs4PuuHtwQHkz1r78FOsJZWIv6FSqFNsYBLANSF5YUkrpDDqsCJqc+NSpB7qZWE5Pw/nqExyPpyWzEBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T20:20:22.462447Z"},"content_sha256":"b6e532350ddf86688d39c15f5dfd0bcdf8a63c84badbcf7e1426e7a741eab30e","schema_version":"1.0","event_id":"sha256:b6e532350ddf86688d39c15f5dfd0bcdf8a63c84badbcf7e1426e7a741eab30e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q7R4TJMRXURKWJ7OQQHR4R7VJO/bundle.json","state_url":"https://pith.science/pith/Q7R4TJMRXURKWJ7OQQHR4R7VJO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q7R4TJMRXURKWJ7OQQHR4R7VJO/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":"2026-08-19T20:20:22Z","links":{"resolver":"https://pith.science/pith/Q7R4TJMRXURKWJ7OQQHR4R7VJO","bundle":"https://pith.science/pith/Q7R4TJMRXURKWJ7OQQHR4R7VJO/bundle.json","state":"https://pith.science/pith/Q7R4TJMRXURKWJ7OQQHR4R7VJO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q7R4TJMRXURKWJ7OQQHR4R7VJO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:Q7R4TJMRXURKWJ7OQQHR4R7VJO","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":"2a0fcd2a430b310d2b5e7995b0a6157b5cad65900ebe416d08cbc62691ad0169","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-28T05:51:45Z","title_canon_sha256":"c91f09613a68ebabdbe39ace951b3f088a36ff7a2a0dc7e1673ee4a18407b2e9"},"schema_version":"1.0","source":{"id":"2411.18932","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.18932","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"arxiv_version","alias_value":"2411.18932v1","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.18932","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"pith_short_12","alias_value":"Q7R4TJMRXURK","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"pith_short_16","alias_value":"Q7R4TJMRXURKWJ7O","created_at":"2026-07-05T09:41:54Z"},{"alias_kind":"pith_short_8","alias_value":"Q7R4TJMR","created_at":"2026-07-05T09:41:54Z"}],"graph_snapshots":[{"event_id":"sha256:b6e532350ddf86688d39c15f5dfd0bcdf8a63c84badbcf7e1426e7a741eab30e","target":"graph","created_at":"2026-07-05T09:41:54Z","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/2411.18932/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in large multimodal models (LMMs) have showcased impressive code generation capabilities, primarily evaluated through image-to-code benchmarks. However, these benchmarks are limited to specific visual programming scenarios where the logic reasoning and the multimodal understanding capacities are split apart. To fill this gap, we propose ScratchEval, a novel benchmark designed to evaluate the visual programming reasoning ability of LMMs. ScratchEval is based on Scratch, a block-based visual programming language widely used in children's programming education. By integrating ","authors_text":"Hongzhan Lin, Jing Ma, Rao Fu, Zhen Ye, Ziyang Luo","cross_cats":["cs.AI","cs.CV"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-28T05:51:45Z","title":"ScratchEval: Are GPT-4o Smarter than My Child? Evaluating Large Multimodal Models with Visual Programming Challenges"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.18932","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:d9a26e63efcb6c91184cd96646607fa2836e1fc20156cc1073928bad49f44a1b","target":"record","created_at":"2026-07-05T09:41:54Z","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":"2a0fcd2a430b310d2b5e7995b0a6157b5cad65900ebe416d08cbc62691ad0169","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-28T05:51:45Z","title_canon_sha256":"c91f09613a68ebabdbe39ace951b3f088a36ff7a2a0dc7e1673ee4a18407b2e9"},"schema_version":"1.0","source":{"id":"2411.18932","kind":"arxiv","version":1}},"canonical_sha256":"87e3c9a591bd22ab27ee840f1e47f54bbe9ea705f6e050781dc2fd678c3677f4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"87e3c9a591bd22ab27ee840f1e47f54bbe9ea705f6e050781dc2fd678c3677f4","first_computed_at":"2026-07-05T09:41:54.146944Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:41:54.146944Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"A6KEkDTpUbd4Hie0qhCJbhHsgzcE0AbcBvZdflo2V/LQBM61x6EQhmK8JDkN9EkU8El2HBpxf9AywbqXUpk4Cg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:41:54.147408Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.18932","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d9a26e63efcb6c91184cd96646607fa2836e1fc20156cc1073928bad49f44a1b","sha256:b6e532350ddf86688d39c15f5dfd0bcdf8a63c84badbcf7e1426e7a741eab30e"],"state_sha256":"2123e39d803b3a6b0910681b201c6cd20e8e0e28c90877cb292379edf8949070"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wOFlbxgJxSGi9lD9Ji6Aai6Z2DIwor4TjHxZaHbW8MhEE5fRlwSpyweGts0en+KAphYbWDDF/zy9be6dwdO/BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T20:20:22.469593Z","bundle_sha256":"30c36149fbe156a0289f4df019894db8460dfeec1f46fb5d69c8314b03a3f5d9"}}