{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:VMFD4SW7EXDC3YPOFLC2GI65ZI","short_pith_number":"pith:VMFD4SW7","canonical_record":{"source":{"id":"2407.01892","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-07-02T02:27:46Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f37c389f4b71664c7c5ed01b88b7eb772ee82c74ec329ce4718ea1748dcf12c2","abstract_canon_sha256":"1feb246491160e2a2a6a716fd1e84271bd5b4fed7c8e4cc6aa76a31b34e65300"},"schema_version":"1.0"},"canonical_sha256":"ab0a3e4adf25c62de1ee2ac5a323ddca09f48ca451dbf06bf466fd0ec64ee086","source":{"kind":"arxiv","id":"2407.01892","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.01892","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"arxiv_version","alias_value":"2407.01892v2","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.01892","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"pith_short_12","alias_value":"VMFD4SW7EXDC","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"pith_short_16","alias_value":"VMFD4SW7EXDC3YPO","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"pith_short_8","alias_value":"VMFD4SW7","created_at":"2026-07-05T10:02:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:VMFD4SW7EXDC3YPOFLC2GI65ZI","target":"record","payload":{"canonical_record":{"source":{"id":"2407.01892","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-07-02T02:27:46Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f37c389f4b71664c7c5ed01b88b7eb772ee82c74ec329ce4718ea1748dcf12c2","abstract_canon_sha256":"1feb246491160e2a2a6a716fd1e84271bd5b4fed7c8e4cc6aa76a31b34e65300"},"schema_version":"1.0"},"canonical_sha256":"ab0a3e4adf25c62de1ee2ac5a323ddca09f48ca451dbf06bf466fd0ec64ee086","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:04.806658Z","signature_b64":"vJ/RhTE3ZAOTJE0bhAA4X5sRDHzzmNuohdasq4F4d6e0U1OVpqYzBidqtxl1juw0NQZV2GXAHIkd9iN3M4ImCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ab0a3e4adf25c62de1ee2ac5a323ddca09f48ca451dbf06bf466fd0ec64ee086","last_reissued_at":"2026-07-05T10:02:04.806213Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:04.806213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.01892","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:02:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ng4d0aqI5HJwpD3d5+m1ZMwlA+OQ6QzU6Qnzlbu+HTDy/KAvNMy16MoBRWqLrQxx2fhoY62xHOuLtrmsDnHvDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T09:17:38.720163Z"},"content_sha256":"b10f48d8c927fe0aef306dc6e67e4c496f41fa64bf5b97450c9dd0ad57760d4e","schema_version":"1.0","event_id":"sha256:b10f48d8c927fe0aef306dc6e67e4c496f41fa64bf5b97450c9dd0ad57760d4e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:VMFD4SW7EXDC3YPOFLC2GI65ZI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GRASP: A Grid-Based Benchmark for Evaluating Commonsense Spatial Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Mayank Kejriwal, Zhisheng Tang","submitted_at":"2024-07-02T02:27:46Z","abstract_excerpt":"Spatial reasoning, an important faculty of human cognition with many practical applications, is one of the core commonsense skills that is not purely language-based and, for satisfying (as opposed to optimal) solutions, requires some minimum degree of planning. Existing benchmarks of Commonsense Spatial Reasoning (CSR) tend to evaluate how Large Language Models (LLMs) interpret text-based spatial $\\textit{descriptions}$ rather than directly evaluate a plan produced by the LLM in response to a $\\textit{specific}$ spatial reasoning problem. In this paper, we construct a large-scale benchmark cal"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.01892","kind":"arxiv","version":2},"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/2407.01892/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":"2026-07-05T10:02:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ne+aTMCDvkpWzKvP6lEwWhNvdTVFETjoKAfLRgEZrKv0XpZZa1l2kF7V5ggVp2aVtt0kzM1VKyH98zWrNZE+Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T09:17:38.720681Z"},"content_sha256":"e92fbae5ba31f2f23772fce9b2fd961d73bb3c2f4806604c9289bbaeb4029121","schema_version":"1.0","event_id":"sha256:e92fbae5ba31f2f23772fce9b2fd961d73bb3c2f4806604c9289bbaeb4029121"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VMFD4SW7EXDC3YPOFLC2GI65ZI/bundle.json","state_url":"https://pith.science/pith/VMFD4SW7EXDC3YPOFLC2GI65ZI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VMFD4SW7EXDC3YPOFLC2GI65ZI/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":"2026-08-18T09:17:38Z","links":{"resolver":"https://pith.science/pith/VMFD4SW7EXDC3YPOFLC2GI65ZI","bundle":"https://pith.science/pith/VMFD4SW7EXDC3YPOFLC2GI65ZI/bundle.json","state":"https://pith.science/pith/VMFD4SW7EXDC3YPOFLC2GI65ZI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VMFD4SW7EXDC3YPOFLC2GI65ZI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:VMFD4SW7EXDC3YPOFLC2GI65ZI","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":"1feb246491160e2a2a6a716fd1e84271bd5b4fed7c8e4cc6aa76a31b34e65300","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-07-02T02:27:46Z","title_canon_sha256":"f37c389f4b71664c7c5ed01b88b7eb772ee82c74ec329ce4718ea1748dcf12c2"},"schema_version":"1.0","source":{"id":"2407.01892","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.01892","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"arxiv_version","alias_value":"2407.01892v2","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.01892","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"pith_short_12","alias_value":"VMFD4SW7EXDC","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"pith_short_16","alias_value":"VMFD4SW7EXDC3YPO","created_at":"2026-07-05T10:02:04Z"},{"alias_kind":"pith_short_8","alias_value":"VMFD4SW7","created_at":"2026-07-05T10:02:04Z"}],"graph_snapshots":[{"event_id":"sha256:e92fbae5ba31f2f23772fce9b2fd961d73bb3c2f4806604c9289bbaeb4029121","target":"graph","created_at":"2026-07-05T10:02:04Z","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/2407.01892/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Spatial reasoning, an important faculty of human cognition with many practical applications, is one of the core commonsense skills that is not purely language-based and, for satisfying (as opposed to optimal) solutions, requires some minimum degree of planning. Existing benchmarks of Commonsense Spatial Reasoning (CSR) tend to evaluate how Large Language Models (LLMs) interpret text-based spatial $\\textit{descriptions}$ rather than directly evaluate a plan produced by the LLM in response to a $\\textit{specific}$ spatial reasoning problem. In this paper, we construct a large-scale benchmark cal","authors_text":"Mayank Kejriwal, Zhisheng Tang","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-07-02T02:27:46Z","title":"GRASP: A Grid-Based Benchmark for Evaluating Commonsense Spatial Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.01892","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:b10f48d8c927fe0aef306dc6e67e4c496f41fa64bf5b97450c9dd0ad57760d4e","target":"record","created_at":"2026-07-05T10:02:04Z","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":"1feb246491160e2a2a6a716fd1e84271bd5b4fed7c8e4cc6aa76a31b34e65300","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-07-02T02:27:46Z","title_canon_sha256":"f37c389f4b71664c7c5ed01b88b7eb772ee82c74ec329ce4718ea1748dcf12c2"},"schema_version":"1.0","source":{"id":"2407.01892","kind":"arxiv","version":2}},"canonical_sha256":"ab0a3e4adf25c62de1ee2ac5a323ddca09f48ca451dbf06bf466fd0ec64ee086","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ab0a3e4adf25c62de1ee2ac5a323ddca09f48ca451dbf06bf466fd0ec64ee086","first_computed_at":"2026-07-05T10:02:04.806213Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:02:04.806213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vJ/RhTE3ZAOTJE0bhAA4X5sRDHzzmNuohdasq4F4d6e0U1OVpqYzBidqtxl1juw0NQZV2GXAHIkd9iN3M4ImCA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:02:04.806658Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.01892","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b10f48d8c927fe0aef306dc6e67e4c496f41fa64bf5b97450c9dd0ad57760d4e","sha256:e92fbae5ba31f2f23772fce9b2fd961d73bb3c2f4806604c9289bbaeb4029121"],"state_sha256":"28b7c6bbe283ff0fc74e78ca29b2f306c0c7713766ae243c32b6a3175ad19138"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D3ZpTuCrMPNXXX4CN3KqxLefJKqfwoKPcXWgV3Rog3tWKcziJXE6uO9uBLso4DQIwAtRdhdpubIUhTjzOJZBCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T09:17:38.725601Z","bundle_sha256":"38d11691c63bded3a9e2b13961b8aef5afe9808d70b48991e2e08d882a2e44df"}}