{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:CIIN6DYSJPL6Y4FA3LA3UU3P2T","short_pith_number":"pith:CIIN6DYS","canonical_record":{"source":{"id":"2405.15064","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T21:22:00Z","cross_cats_sorted":["cs.AI","cs.DB"],"title_canon_sha256":"861256f9c12d84924ed3100c41569192ce119b33b119ae9acb7483001014dcc6","abstract_canon_sha256":"8937c16ccceed2697ae04bc2e4a4f39e7b7e6815c0041b1790f235c819cc2a89"},"schema_version":"1.0"},"canonical_sha256":"1210df0f124bd7ec70a0dac1ba536fd4c67e5b3822126bae3aea65c975c3a436","source":{"kind":"arxiv","id":"2405.15064","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.15064","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"arxiv_version","alias_value":"2405.15064v1","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.15064","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"pith_short_12","alias_value":"CIIN6DYSJPL6","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"pith_short_16","alias_value":"CIIN6DYSJPL6Y4FA","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"pith_short_8","alias_value":"CIIN6DYS","created_at":"2026-07-05T08:22:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:CIIN6DYSJPL6Y4FA3LA3UU3P2T","target":"record","payload":{"canonical_record":{"source":{"id":"2405.15064","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T21:22:00Z","cross_cats_sorted":["cs.AI","cs.DB"],"title_canon_sha256":"861256f9c12d84924ed3100c41569192ce119b33b119ae9acb7483001014dcc6","abstract_canon_sha256":"8937c16ccceed2697ae04bc2e4a4f39e7b7e6815c0041b1790f235c819cc2a89"},"schema_version":"1.0"},"canonical_sha256":"1210df0f124bd7ec70a0dac1ba536fd4c67e5b3822126bae3aea65c975c3a436","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:22:44.692850Z","signature_b64":"pthsacfuWPwk0E9efwNf/3Vb+WZwvZinLb0/2UG99TySVPPOHqLKakQUurGz0+k/Uu8l2UH5wUIBzvlhB5AaDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1210df0f124bd7ec70a0dac1ba536fd4c67e5b3822126bae3aea65c975c3a436","last_reissued_at":"2026-07-05T08:22:44.692399Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:22:44.692399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.15064","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":"2026-07-05T08:22:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ItjAXHSNmKtuKPWxt1+Xa1EQCSQr8onmY79cY+JihjVueaMFttrldH0czbI9foGq9aCTTWN7UrS7paJ6+jPSDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:53:49.744781Z"},"content_sha256":"b5b536bcea0a8ae9db671d96da6c29bf030f836c117641eca2cb0960f4b263db","schema_version":"1.0","event_id":"sha256:b5b536bcea0a8ae9db671d96da6c29bf030f836c117641eca2cb0960f4b263db"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:CIIN6DYSJPL6Y4FA3LA3UU3P2T","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reframing Spatial Reasoning Evaluation in Language Models: A Real-World Simulation Benchmark for Qualitative Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DB"],"primary_cat":"cs.CL","authors_text":"Anthony G. Cohn, David C. Hogg, Fangjun Li","submitted_at":"2024-05-23T21:22:00Z","abstract_excerpt":"Spatial reasoning plays a vital role in both human cognition and machine intelligence, prompting new research into language models' (LMs) capabilities in this regard. However, existing benchmarks reveal shortcomings in evaluating qualitative spatial reasoning (QSR). These benchmarks typically present oversimplified scenarios or unclear natural language descriptions, hindering effective evaluation. We present a novel benchmark for assessing QSR in LMs, which is grounded in realistic 3D simulation data, offering a series of diverse room layouts with various objects and their spatial relationship"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.15064","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/2405.15064/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":"2026-07-05T08:22:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KcuYtO8a5w/0qAtag9OMpsem4Zsc48jI16iXwkgcCulJ4JSCHBCWk5idDnLrYeH1/PLOLBVniZiX43rMD5ncCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:53:49.745380Z"},"content_sha256":"13c24fac086d8e94e9512379e352a27ff416b0bcfe0e0efce6652927287c6cdf","schema_version":"1.0","event_id":"sha256:13c24fac086d8e94e9512379e352a27ff416b0bcfe0e0efce6652927287c6cdf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CIIN6DYSJPL6Y4FA3LA3UU3P2T/bundle.json","state_url":"https://pith.science/pith/CIIN6DYSJPL6Y4FA3LA3UU3P2T/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CIIN6DYSJPL6Y4FA3LA3UU3P2T/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":"2026-08-08T13:53:49Z","links":{"resolver":"https://pith.science/pith/CIIN6DYSJPL6Y4FA3LA3UU3P2T","bundle":"https://pith.science/pith/CIIN6DYSJPL6Y4FA3LA3UU3P2T/bundle.json","state":"https://pith.science/pith/CIIN6DYSJPL6Y4FA3LA3UU3P2T/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CIIN6DYSJPL6Y4FA3LA3UU3P2T/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CIIN6DYSJPL6Y4FA3LA3UU3P2T","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":"8937c16ccceed2697ae04bc2e4a4f39e7b7e6815c0041b1790f235c819cc2a89","cross_cats_sorted":["cs.AI","cs.DB"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T21:22:00Z","title_canon_sha256":"861256f9c12d84924ed3100c41569192ce119b33b119ae9acb7483001014dcc6"},"schema_version":"1.0","source":{"id":"2405.15064","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.15064","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"arxiv_version","alias_value":"2405.15064v1","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.15064","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"pith_short_12","alias_value":"CIIN6DYSJPL6","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"pith_short_16","alias_value":"CIIN6DYSJPL6Y4FA","created_at":"2026-07-05T08:22:44Z"},{"alias_kind":"pith_short_8","alias_value":"CIIN6DYS","created_at":"2026-07-05T08:22:44Z"}],"graph_snapshots":[{"event_id":"sha256:13c24fac086d8e94e9512379e352a27ff416b0bcfe0e0efce6652927287c6cdf","target":"graph","created_at":"2026-07-05T08:22:44Z","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/2405.15064/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Spatial reasoning plays a vital role in both human cognition and machine intelligence, prompting new research into language models' (LMs) capabilities in this regard. However, existing benchmarks reveal shortcomings in evaluating qualitative spatial reasoning (QSR). These benchmarks typically present oversimplified scenarios or unclear natural language descriptions, hindering effective evaluation. We present a novel benchmark for assessing QSR in LMs, which is grounded in realistic 3D simulation data, offering a series of diverse room layouts with various objects and their spatial relationship","authors_text":"Anthony G. Cohn, David C. Hogg, Fangjun Li","cross_cats":["cs.AI","cs.DB"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T21:22:00Z","title":"Reframing Spatial Reasoning Evaluation in Language Models: A Real-World Simulation Benchmark for Qualitative Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.15064","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:b5b536bcea0a8ae9db671d96da6c29bf030f836c117641eca2cb0960f4b263db","target":"record","created_at":"2026-07-05T08:22:44Z","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":"8937c16ccceed2697ae04bc2e4a4f39e7b7e6815c0041b1790f235c819cc2a89","cross_cats_sorted":["cs.AI","cs.DB"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-23T21:22:00Z","title_canon_sha256":"861256f9c12d84924ed3100c41569192ce119b33b119ae9acb7483001014dcc6"},"schema_version":"1.0","source":{"id":"2405.15064","kind":"arxiv","version":1}},"canonical_sha256":"1210df0f124bd7ec70a0dac1ba536fd4c67e5b3822126bae3aea65c975c3a436","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1210df0f124bd7ec70a0dac1ba536fd4c67e5b3822126bae3aea65c975c3a436","first_computed_at":"2026-07-05T08:22:44.692399Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:22:44.692399Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pthsacfuWPwk0E9efwNf/3Vb+WZwvZinLb0/2UG99TySVPPOHqLKakQUurGz0+k/Uu8l2UH5wUIBzvlhB5AaDw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:22:44.692850Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.15064","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b5b536bcea0a8ae9db671d96da6c29bf030f836c117641eca2cb0960f4b263db","sha256:13c24fac086d8e94e9512379e352a27ff416b0bcfe0e0efce6652927287c6cdf"],"state_sha256":"ea261bea422d263a8abb921cb814b10e98af3c5e72792dbce2356be97e3cdc66"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"17aLBlcFdhDB4G7kTebZFydzqUuhKeUe3qLWpNarmy+5P1PfJMvix4fhrMWkGdQcjYceLe6460iD2inc8OZGDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T13:53:49.749371Z","bundle_sha256":"b2b28a9ee6d0d0421a9dc010fd3ad62aac1bf9cf3c315dbaa18851baf64e12e9"}}