{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OIKLPDFWNA7E2WK4TMLGB2TUP5","short_pith_number":"pith:OIKLPDFW","schema_version":"1.0","canonical_sha256":"7214b78cb6683e4d595c9b1660ea747f4bf2f6aeda9141c1883727ce5dfed45e","source":{"kind":"arxiv","id":"2504.12110","version":2},"attestation_state":"computed","paper":{"title":"Towards LLM Agents for Earth Observation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bharath Hariharan, Carl Vondrick, Cheng Perng Phoo, Chia Hsiang Kao, Kavita Bala, Rajeev Datta, Samuel Speas, Shreelekha Revankar, Snehal Bhagat, Utkarsh Mall, Wenting Zhao","submitted_at":"2025-04-16T14:19:25Z","abstract_excerpt":"Earth Observation (EO) provides critical planetary data for environmental monitoring, disaster management, climate science, and other scientific domains. Here we ask: Are AI systems ready for reliable Earth Observation? We introduce \\datasetnamenospace, a benchmark of 140 yes/no questions from NASA Earth Observatory articles across 13 topics and 17 satellite sensors. Using Google Earth Engine API as a tool, LLM agents can only achieve an accuracy of 33% because the code fails to run over 58% of the time. We improve the failure rate for open models by fine-tuning synthetic data, allowing much s"},"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":"2504.12110","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-16T14:19:25Z","cross_cats_sorted":[],"title_canon_sha256":"a45439b53200eead007207aa6c0b2bb524392ea1318240ff49dba67ee6f9e552","abstract_canon_sha256":"33811b8fd1893f0e8f282a77e3f47406276b1409d0a9c195794431a7e7fa9f3e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:17.106280Z","signature_b64":"uLKN83jI0lh7xQR5RbNwgq5ZFOXpvTTDLPOorqkr2wUYvSYe2WDK+YJZNcCwHqS+vCBEiYYK6Nom7xR5G3QWAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7214b78cb6683e4d595c9b1660ea747f4bf2f6aeda9141c1883727ce5dfed45e","last_reissued_at":"2026-07-05T12:11:17.105640Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:17.105640Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards LLM Agents for Earth Observation","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bharath Hariharan, Carl Vondrick, Cheng Perng Phoo, Chia Hsiang Kao, Kavita Bala, Rajeev Datta, Samuel Speas, Shreelekha Revankar, Snehal Bhagat, Utkarsh Mall, Wenting Zhao","submitted_at":"2025-04-16T14:19:25Z","abstract_excerpt":"Earth Observation (EO) provides critical planetary data for environmental monitoring, disaster management, climate science, and other scientific domains. Here we ask: Are AI systems ready for reliable Earth Observation? We introduce \\datasetnamenospace, a benchmark of 140 yes/no questions from NASA Earth Observatory articles across 13 topics and 17 satellite sensors. Using Google Earth Engine API as a tool, LLM agents can only achieve an accuracy of 33% because the code fails to run over 58% of the time. We improve the failure rate for open models by fine-tuning synthetic data, allowing much s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.12110","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/2504.12110/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":"2504.12110","created_at":"2026-07-05T12:11:17.105739+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.12110v2","created_at":"2026-07-05T12:11:17.105739+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.12110","created_at":"2026-07-05T12:11:17.105739+00:00"},{"alias_kind":"pith_short_12","alias_value":"OIKLPDFWNA7E","created_at":"2026-07-05T12:11:17.105739+00:00"},{"alias_kind":"pith_short_16","alias_value":"OIKLPDFWNA7E2WK4","created_at":"2026-07-05T12:11:17.105739+00:00"},{"alias_kind":"pith_short_8","alias_value":"OIKLPDFW","created_at":"2026-07-05T12:11:17.105739+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13148","citing_title":"TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13148","citing_title":"TerraBench: Can Agents Reason Over Heterogeneous Earth-System Data?","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12821","citing_title":"GeoNatureAgent Benchmark: Benchmarking LLM Agents for Environmental Geospatial Analysis Across Frontier and Open-Weight Foundation Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24919","citing_title":"Agentic AI for Remote Sensing: Technical Challenges and Research Directions","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11633","citing_title":"Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24919","citing_title":"Agentic AI for Remote Sensing: Technical Challenges and Research Directions","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01250","citing_title":"EO-Gym: A Multimodal, Interactive Environment for Earth Observation Agents","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5","json":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5.json","graph_json":"https://pith.science/api/pith-number/OIKLPDFWNA7E2WK4TMLGB2TUP5/graph.json","events_json":"https://pith.science/api/pith-number/OIKLPDFWNA7E2WK4TMLGB2TUP5/events.json","paper":"https://pith.science/paper/OIKLPDFW"},"agent_actions":{"view_html":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5","download_json":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5.json","view_paper":"https://pith.science/paper/OIKLPDFW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.12110&json=true","fetch_graph":"https://pith.science/api/pith-number/OIKLPDFWNA7E2WK4TMLGB2TUP5/graph.json","fetch_events":"https://pith.science/api/pith-number/OIKLPDFWNA7E2WK4TMLGB2TUP5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5/action/storage_attestation","attest_author":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5/action/author_attestation","sign_citation":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5/action/citation_signature","submit_replication":"https://pith.science/pith/OIKLPDFWNA7E2WK4TMLGB2TUP5/action/replication_record"}},"created_at":"2026-07-05T12:11:17.105739+00:00","updated_at":"2026-07-05T12:11:17.105739+00:00"}