{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ZN7WDERCLDIQKKLDHBMP6XLWFP","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":"dab2b5271da02d83cca25e7fc0f3261b295470751523f24b1d6f3a21e51b2c52","cross_cats_sorted":["cs.CL","cs.IR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T16:58:04Z","title_canon_sha256":"5c759b471dc626b5d2c9534431c059a0cf10a5d36b6e22ca3f08d1972b0d3451"},"schema_version":"1.0","source":{"id":"2608.07411","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.07411","created_at":"2026-08-10T01:15:17Z"},{"alias_kind":"arxiv_version","alias_value":"2608.07411v1","created_at":"2026-08-10T01:15:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07411","created_at":"2026-08-10T01:15:17Z"},{"alias_kind":"pith_short_12","alias_value":"ZN7WDERCLDIQ","created_at":"2026-08-10T01:15:17Z"},{"alias_kind":"pith_short_16","alias_value":"ZN7WDERCLDIQKKLD","created_at":"2026-08-10T01:15:17Z"},{"alias_kind":"pith_short_8","alias_value":"ZN7WDERC","created_at":"2026-08-10T01:15:17Z"}],"graph_snapshots":[{"event_id":"sha256:373d9b1f56ed2025456e480d1ebe62190bb8b5a0d0cdbae9bb492ef360263db5","target":"graph","created_at":"2026-08-10T01:15:17Z","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/2608.07411/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In the context of geodata, existing Large Language Models have often been studied in a homogeneous setting, which has considerably limited insights into their generalization capabilities. In this paper, we present \\benchName, a comprehensive benchmark for probing LLMs on geo-related tasks. We leverage a careful selection of twelve publicly available datasets from diverse geo-related tasks and domains, and evaluate a set of LLMs on geo-spatial and temporal understanding using our benchmark. Our results show that reasoning and size have a strong impact on overall performance. GeoBenchLLM is publ","authors_text":"Jose G Moreno, Karim Radouane, Lynda Tamine, Rodrigo Ferreira Rodrigues","cross_cats":["cs.CL","cs.IR","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T16:58:04Z","title":"GeoBenchLLM: A Comprehensive Benchmark for Evaluating LLMs on Geo-Related Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07411","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":"sha256:584fce52d7128af27e44a1d7ad9775b1bc6ba217cc72302d9c27259ed0513393","target":"record","created_at":"2026-08-10T01:15:17Z","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":"dab2b5271da02d83cca25e7fc0f3261b295470751523f24b1d6f3a21e51b2c52","cross_cats_sorted":["cs.CL","cs.IR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T16:58:04Z","title_canon_sha256":"5c759b471dc626b5d2c9534431c059a0cf10a5d36b6e22ca3f08d1972b0d3451"},"schema_version":"1.0","source":{"id":"2608.07411","kind":"arxiv","version":1}},"canonical_sha256":"cb7f61922258d10529633858ff5d762bf4f16d1d2bb19c1ddbeaf86eeaba6c41","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cb7f61922258d10529633858ff5d762bf4f16d1d2bb19c1ddbeaf86eeaba6c41","first_computed_at":"2026-08-10T01:15:17.996756Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-10T01:15:17.996756Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"THUhJqx3oNsKepznsv15LSVzTFOe35sEHQ/IhVxFwapzp6FbzHlDoODXfz5+gyi1i11PsSZl6dynnu6CeTq1Bg==","signature_status":"signed_v1","signed_at":"2026-08-10T01:15:17.999210Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.07411","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:584fce52d7128af27e44a1d7ad9775b1bc6ba217cc72302d9c27259ed0513393","sha256:373d9b1f56ed2025456e480d1ebe62190bb8b5a0d0cdbae9bb492ef360263db5"],"state_sha256":"274100ecd4d165edf3969c6513a40952e19502f024c2e3bd3a850b6ac3e54184"}