{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:QLTVUHNXJQSN3SGJD7AJ64F7IR","short_pith_number":"pith:QLTVUHNX","canonical_record":{"source":{"id":"2507.20395","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-07-27T19:33:45Z","cross_cats_sorted":[],"title_canon_sha256":"cebb27bb8b0b129e09e16765fddf6dec9bd5711400244bcc8f275a7a346df2e5","abstract_canon_sha256":"f1403f0f18eb17db9d7d50d5172f51d01937a430bf3f6fbfaab7c8c0f64bbb71"},"schema_version":"1.0"},"canonical_sha256":"82e75a1db74c24ddc8c91fc09f70bf44708f105ea3bbcc694a37b67e9e2e6911","source":{"kind":"arxiv","id":"2507.20395","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.20395","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"arxiv_version","alias_value":"2507.20395v1","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.20395","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"pith_short_12","alias_value":"QLTVUHNXJQSN","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"pith_short_16","alias_value":"QLTVUHNXJQSN3SGJ","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"pith_short_8","alias_value":"QLTVUHNX","created_at":"2026-07-05T11:44:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:QLTVUHNXJQSN3SGJD7AJ64F7IR","target":"record","payload":{"canonical_record":{"source":{"id":"2507.20395","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-07-27T19:33:45Z","cross_cats_sorted":[],"title_canon_sha256":"cebb27bb8b0b129e09e16765fddf6dec9bd5711400244bcc8f275a7a346df2e5","abstract_canon_sha256":"f1403f0f18eb17db9d7d50d5172f51d01937a430bf3f6fbfaab7c8c0f64bbb71"},"schema_version":"1.0"},"canonical_sha256":"82e75a1db74c24ddc8c91fc09f70bf44708f105ea3bbcc694a37b67e9e2e6911","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:44:18.532163Z","signature_b64":"D274oUImpAoJ3Zp619at/GfccOOdu4in2ax8U8J/8CJVzbkZH5ZxERDpag88/OJRgd0gCVlSOOj69m99S+ZxAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"82e75a1db74c24ddc8c91fc09f70bf44708f105ea3bbcc694a37b67e9e2e6911","last_reissued_at":"2026-07-05T11:44:18.531708Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:44:18.531708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.20395","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":"2026-07-05T11:44:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d0pYWkU5rdLlwsdsR4U+TxgNkbwWSmTaHVa9P3bF/JDnXz/DbBn2JwVQrScitFoGw4Aj9rTaT1UvZB4zbNQAAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T01:11:46.033974Z"},"content_sha256":"97a71250f89f6fb3dc613af41c2db5e4b67a315f9f436c1b8bf68ad3651e16a7","schema_version":"1.0","event_id":"sha256:97a71250f89f6fb3dc613af41c2db5e4b67a315f9f436c1b8bf68ad3651e16a7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:QLTVUHNXJQSN3SGJD7AJ64F7IR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Hafsteinn Einarsson","submitted_at":"2025-07-27T19:33:45Z","abstract_excerpt":"As Large Language Models (LLMs) increasingly power autonomous agents in robotics and embodied AI, understanding their spatial reasoning capabilities becomes crucial for ensuring reliable real-world deployment. Despite advances in language understanding, current research lacks evaluation of how LLMs perform spatial navigation without visual cues, a fundamental requirement for agents operating with limited sensory information. This paper addresses this gap by introducing MazeEval, a benchmark designed to isolate and evaluate pure spatial reasoning in LLMs through coordinate-based maze navigation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.20395","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/2507.20395/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":"2026-07-05T11:44:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7PHm52jBQVD6SR7uJe/VUGEVE41Qa0j4gCEc55pWysnXv13tb6W6yAXRp5KbiWq0obJsqlS5quo33Vn3u00DCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T01:11:46.034634Z"},"content_sha256":"0cfdbe511d4ee117e92ed4f4f857c64ce32d3672e3ab938fe0012b573955355e","schema_version":"1.0","event_id":"sha256:0cfdbe511d4ee117e92ed4f4f857c64ce32d3672e3ab938fe0012b573955355e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QLTVUHNXJQSN3SGJD7AJ64F7IR/bundle.json","state_url":"https://pith.science/pith/QLTVUHNXJQSN3SGJD7AJ64F7IR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QLTVUHNXJQSN3SGJD7AJ64F7IR/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":"2026-08-09T01:11:46Z","links":{"resolver":"https://pith.science/pith/QLTVUHNXJQSN3SGJD7AJ64F7IR","bundle":"https://pith.science/pith/QLTVUHNXJQSN3SGJD7AJ64F7IR/bundle.json","state":"https://pith.science/pith/QLTVUHNXJQSN3SGJD7AJ64F7IR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QLTVUHNXJQSN3SGJD7AJ64F7IR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:QLTVUHNXJQSN3SGJD7AJ64F7IR","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":"f1403f0f18eb17db9d7d50d5172f51d01937a430bf3f6fbfaab7c8c0f64bbb71","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-07-27T19:33:45Z","title_canon_sha256":"cebb27bb8b0b129e09e16765fddf6dec9bd5711400244bcc8f275a7a346df2e5"},"schema_version":"1.0","source":{"id":"2507.20395","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.20395","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"arxiv_version","alias_value":"2507.20395v1","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.20395","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"pith_short_12","alias_value":"QLTVUHNXJQSN","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"pith_short_16","alias_value":"QLTVUHNXJQSN3SGJ","created_at":"2026-07-05T11:44:18Z"},{"alias_kind":"pith_short_8","alias_value":"QLTVUHNX","created_at":"2026-07-05T11:44:18Z"}],"graph_snapshots":[{"event_id":"sha256:0cfdbe511d4ee117e92ed4f4f857c64ce32d3672e3ab938fe0012b573955355e","target":"graph","created_at":"2026-07-05T11:44:18Z","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/2507.20395/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As Large Language Models (LLMs) increasingly power autonomous agents in robotics and embodied AI, understanding their spatial reasoning capabilities becomes crucial for ensuring reliable real-world deployment. Despite advances in language understanding, current research lacks evaluation of how LLMs perform spatial navigation without visual cues, a fundamental requirement for agents operating with limited sensory information. This paper addresses this gap by introducing MazeEval, a benchmark designed to isolate and evaluate pure spatial reasoning in LLMs through coordinate-based maze navigation","authors_text":"Hafsteinn Einarsson","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-07-27T19:33:45Z","title":"MazeEval: A Benchmark for Testing Sequential Decision-Making in Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.20395","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":"sha256:97a71250f89f6fb3dc613af41c2db5e4b67a315f9f436c1b8bf68ad3651e16a7","target":"record","created_at":"2026-07-05T11:44:18Z","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":"f1403f0f18eb17db9d7d50d5172f51d01937a430bf3f6fbfaab7c8c0f64bbb71","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-07-27T19:33:45Z","title_canon_sha256":"cebb27bb8b0b129e09e16765fddf6dec9bd5711400244bcc8f275a7a346df2e5"},"schema_version":"1.0","source":{"id":"2507.20395","kind":"arxiv","version":1}},"canonical_sha256":"82e75a1db74c24ddc8c91fc09f70bf44708f105ea3bbcc694a37b67e9e2e6911","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"82e75a1db74c24ddc8c91fc09f70bf44708f105ea3bbcc694a37b67e9e2e6911","first_computed_at":"2026-07-05T11:44:18.531708Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:44:18.531708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"D274oUImpAoJ3Zp619at/GfccOOdu4in2ax8U8J/8CJVzbkZH5ZxERDpag88/OJRgd0gCVlSOOj69m99S+ZxAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:44:18.532163Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.20395","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:97a71250f89f6fb3dc613af41c2db5e4b67a315f9f436c1b8bf68ad3651e16a7","sha256:0cfdbe511d4ee117e92ed4f4f857c64ce32d3672e3ab938fe0012b573955355e"],"state_sha256":"5b108302b2dd7a92ca203302833d2de8144d63c6a15a1f5e1b5fb92c3a2c6092"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JXwxlvTCw4bgSYElCh7N35D9bhlK6tGd+/kDc02vB/hb/8dHe/OBSJb0TG6x0YKq2VyPn2IJYI96M0pwjOQODw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T01:11:46.038830Z","bundle_sha256":"6865362580f5bed13af11b5c2ed7d00760352a5145709761b2addbd1c6dcf35c"}}