{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:4F2FWASCOQZUJWICTVSUG36LUC","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":"85ca4c4ad6018e611e6bfaa057b932377c40f3b45ccc44beb5ade69e555d4233","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-16T08:52:27Z","title_canon_sha256":"70395720768a745ebabfff7277c71f3d6a848fad743ad7c42d5bb98002bf0443"},"schema_version":"1.0","source":{"id":"2311.09682","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.09682","created_at":"2026-07-05T10:18:33Z"},{"alias_kind":"arxiv_version","alias_value":"2311.09682v4","created_at":"2026-07-05T10:18:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.09682","created_at":"2026-07-05T10:18:33Z"},{"alias_kind":"pith_short_12","alias_value":"4F2FWASCOQZU","created_at":"2026-07-05T10:18:33Z"},{"alias_kind":"pith_short_16","alias_value":"4F2FWASCOQZUJWIC","created_at":"2026-07-05T10:18:33Z"},{"alias_kind":"pith_short_8","alias_value":"4F2FWASC","created_at":"2026-07-05T10:18:33Z"}],"graph_snapshots":[{"event_id":"sha256:9063e4bf7465647592e543ea332208879da7367450c3eb1b1f752e73a28f5847","target":"graph","created_at":"2026-07-05T10:18:33Z","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/2311.09682/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We explore the creative problem-solving capabilities of modern LLMs in a novel constrained setting. To this end, we create MACGYVER, an automatically generated dataset consisting of over 1,600 real-world problems deliberately designed to trigger innovative usage of objects and necessitate out-of-the-box thinking. We then present our collection to both LLMs and humans to compare and contrast their problem-solving abilities. MACGYVER is challenging for both groups, but in unique and complementary ways. For instance, humans excel in tasks they are familiar with but struggle with domain-specific k","authors_text":"Abhilasha Ravichander, Faeze Brahman, Lianhui Qin, Nanyun Peng, Raja Marjieh, Ronan Le Bras, Thomas L. Griffiths, Yejin Choi, Yufei Tian","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-16T08:52:27Z","title":"MacGyver: Are Large Language Models Creative Problem Solvers?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.09682","kind":"arxiv","version":4},"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":"sha256:49a2cc9142e5a6f28d095a72ec9c19815622a05885fd2918bc4094b6a557b578","target":"record","created_at":"2026-07-05T10:18:33Z","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":"85ca4c4ad6018e611e6bfaa057b932377c40f3b45ccc44beb5ade69e555d4233","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-16T08:52:27Z","title_canon_sha256":"70395720768a745ebabfff7277c71f3d6a848fad743ad7c42d5bb98002bf0443"},"schema_version":"1.0","source":{"id":"2311.09682","kind":"arxiv","version":4}},"canonical_sha256":"e1745b0242743344d9029d65436fcba0b00905c2b131e3c4ba15dec9aedeadc2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e1745b0242743344d9029d65436fcba0b00905c2b131e3c4ba15dec9aedeadc2","first_computed_at":"2026-07-05T10:18:33.681912Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:18:33.681912Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8RC2LHsoWUcOEgGp2L94UCw+RXPUAGcN28vYk9p6PF15aF+XKJnoJUggmbw7J0oocypLbv1uDigH2noNnibKAA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:18:33.682415Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.09682","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:49a2cc9142e5a6f28d095a72ec9c19815622a05885fd2918bc4094b6a557b578","sha256:9063e4bf7465647592e543ea332208879da7367450c3eb1b1f752e73a28f5847"],"state_sha256":"c23d0155114855c528694e47dc7705e6501363ab21b64b27c7974e46ce089636"}