{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5644WYU7Y7DRBODX4ZFMYEO7GQ","short_pith_number":"pith:5644WYU7","schema_version":"1.0","canonical_sha256":"efb9cb629fc7c710b877e64acc11df3432c5d17963bc1a05383c3995304f7632","source":{"kind":"arxiv","id":"2607.22732","version":1},"attestation_state":"computed","paper":{"title":"Spatial Reasoning in LLM Game Agents: Impact of Causal Context and Multi-Step Planning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alexander Dockhorn, Bodo Rosenhahn, Mohit Jiwatode, Robin Schm\\\"ocker, Ronja Fuchs","submitted_at":"2026-07-22T12:10:45Z","abstract_excerpt":"LLM-based game agents often perform poorly on more complex tasks. This work examines whether these failures are linked to limited spatial reasoning and evaluates whether causal prompt augmentation and multi-step planning can improve win-rates while managing response latency. Using the open-source Qwen3 model family, we conduct experiments across varying model scales, reasoning modes, and planning horizons. We further introduce a focused GVGAI benchmark consisting of three custom games with five difficulty levels to isolate spatial navigation. The evaluation follows two paradigms: an initial ``"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2607.22732","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-22T12:10:45Z","cross_cats_sorted":[],"title_canon_sha256":"fe894116f3efad0b6af1f4884b9101adb4d243913f98cfd773729ef656f5f27a","abstract_canon_sha256":"9a53e08640f2a47e1bd8dd42a50672577e871c54dca491e7585b33f3841e8509"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T00:21:51.621466Z","signature_b64":"bKxVfV36ODBdOArLTt7xxXqAF75rROzTcTVQLDD/+dsDVS1saH/8bUYd224F20HCqMkMAopbdYiwB+GyRYmlCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"efb9cb629fc7c710b877e64acc11df3432c5d17963bc1a05383c3995304f7632","last_reissued_at":"2026-07-28T00:21:51.620601Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T00:21:51.620601Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Spatial Reasoning in LLM Game Agents: Impact of Causal Context and Multi-Step Planning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Alexander Dockhorn, Bodo Rosenhahn, Mohit Jiwatode, Robin Schm\\\"ocker, Ronja Fuchs","submitted_at":"2026-07-22T12:10:45Z","abstract_excerpt":"LLM-based game agents often perform poorly on more complex tasks. This work examines whether these failures are linked to limited spatial reasoning and evaluates whether causal prompt augmentation and multi-step planning can improve win-rates while managing response latency. Using the open-source Qwen3 model family, we conduct experiments across varying model scales, reasoning modes, and planning horizons. We further introduce a focused GVGAI benchmark consisting of three custom games with five difficulty levels to isolate spatial navigation. The evaluation follows two paradigms: an initial ``"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22732","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/2607.22732/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2607.22732","created_at":"2026-07-28T00:21:51.621035+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.22732v1","created_at":"2026-07-28T00:21:51.621035+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22732","created_at":"2026-07-28T00:21:51.621035+00:00"},{"alias_kind":"pith_short_12","alias_value":"5644WYU7Y7DR","created_at":"2026-07-28T00:21:51.621035+00:00"},{"alias_kind":"pith_short_16","alias_value":"5644WYU7Y7DRBODX","created_at":"2026-07-28T00:21:51.621035+00:00"},{"alias_kind":"pith_short_8","alias_value":"5644WYU7","created_at":"2026-07-28T00:21:51.621035+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ","json":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ.json","graph_json":"https://pith.science/api/pith-number/5644WYU7Y7DRBODX4ZFMYEO7GQ/graph.json","events_json":"https://pith.science/api/pith-number/5644WYU7Y7DRBODX4ZFMYEO7GQ/events.json","paper":"https://pith.science/paper/5644WYU7"},"agent_actions":{"view_html":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ","download_json":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ.json","view_paper":"https://pith.science/paper/5644WYU7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.22732&json=true","fetch_graph":"https://pith.science/api/pith-number/5644WYU7Y7DRBODX4ZFMYEO7GQ/graph.json","fetch_events":"https://pith.science/api/pith-number/5644WYU7Y7DRBODX4ZFMYEO7GQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ/action/storage_attestation","attest_author":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ/action/author_attestation","sign_citation":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ/action/citation_signature","submit_replication":"https://pith.science/pith/5644WYU7Y7DRBODX4ZFMYEO7GQ/action/replication_record"}},"created_at":"2026-07-28T00:21:51.621035+00:00","updated_at":"2026-07-28T00:21:51.621035+00:00"}