{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:LGXGNG4SNSXU7ZUSQBC7S37F2D","short_pith_number":"pith:LGXGNG4S","canonical_record":{"source":{"id":"2605.14120","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T21:11:14Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"63d3eaa1aff67688201a21e2e89160866d6a927959f158422f7598e99902f47e","abstract_canon_sha256":"ade8fde49fb7ece80554c13675da1ea7c1fe6bde903b0dab5674383cc26aea8b"},"schema_version":"1.0"},"canonical_sha256":"59ae669b926caf4fe6928045f96fe5d0d67cb6b15e4cbc6c1077e9316f8b229d","source":{"kind":"arxiv","id":"2605.14120","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14120","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14120v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14120","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"LGXGNG4SNSXU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"LGXGNG4SNSXU7ZUS","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"LGXGNG4S","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:LGXGNG4SNSXU7ZUSQBC7S37F2D","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14120","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T21:11:14Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"63d3eaa1aff67688201a21e2e89160866d6a927959f158422f7598e99902f47e","abstract_canon_sha256":"ade8fde49fb7ece80554c13675da1ea7c1fe6bde903b0dab5674383cc26aea8b"},"schema_version":"1.0"},"canonical_sha256":"59ae669b926caf4fe6928045f96fe5d0d67cb6b15e4cbc6c1077e9316f8b229d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:11.905969Z","signature_b64":"yvCIhv6VmSAVi0T4hFHyvttbPYg34eJx75D7fzSzNnmCJSshVN0S3ebLd9Ysk8BsRlXuGeoun++oLg41RZeaBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59ae669b926caf4fe6928045f96fe5d0d67cb6b15e4cbc6c1077e9316f8b229d","last_reissued_at":"2026-05-17T23:39:11.905033Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:11.905033Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14120","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":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"s8Hgx0A6aGXf7lpg3s3/8Y1bJJlQQyKr5sf99D2fi35Dp+ys0euFNj4E478jP0vzKct4FosvbEd9CdpRJqj3DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-18T17:25:11.752310Z"},"content_sha256":"b75473526b2af0c8318db66181b42f5eb4ad04d4129ca39adb6466ac3dd2c3bc","schema_version":"1.0","event_id":"sha256:b75473526b2af0c8318db66181b42f5eb4ad04d4129ca39adb6466ac3dd2c3bc"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:LGXGNG4SNSXU7ZUSQBC7S37F2D","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Mini-JEPA Foundation Model Fleet Enables Agentic Hydrologic Intelligence","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A fleet of five small specialized foundation models routed by an LLM improves hydrologic reasoning over a single generalist model.","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Mashrekur Rahman","submitted_at":"2026-05-13T21:11:14Z","abstract_excerpt":"Geospatial foundation models compress multispectral observations into dense embeddings increasingly used in natural-language environmental reasoning systems. A single planetary-scale model, e.g. Google AlphaEarth, handles broad characterization well but may compromise on specialized hydrologic signals. Such generalist models are also often inaccessible, expensive, and require large-scale compute. We propose Mini-JEPAs: a fleet of small sensor-specialized Joint Embedding Predictive Architecture (JEPA) foundation models consulted by a routing agent for specialized questions. We pretrained five 2"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"In paired LLM-as-Judge evaluation, dual retrieval over AlphaEarth and the routed fleet outperforms AlphaEarth alone on physics-matched questions (Cohen's d = 1.10, p = 0.031).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the five Mini-JEPA manifolds capture hydrologic signals genuinely absent from AlphaEarth and that the router LLM's perfect hit rate on the curated question set will hold for real user queries outside the test distribution.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A fleet of sensor-specialized 22M-parameter JEPA models routed by an LLM improves LLM-as-judge scores on hydrologic questions over AlphaEarth alone with Cohen's d of 1.10.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A fleet of five small specialized foundation models routed by an LLM improves hydrologic reasoning over a single generalist model.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4a99564c04f01b4a593cee2be331f9bd447afbbedc40185a5a10cf772e12db79"},"source":{"id":"2605.14120","kind":"arxiv","version":1},"verdict":{"id":"d54fdc04-6275-4eb2-b9b5-bb375ffbf311","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T04:57:27.489456Z","strongest_claim":"In paired LLM-as-Judge evaluation, dual retrieval over AlphaEarth and the routed fleet outperforms AlphaEarth alone on physics-matched questions (Cohen's d = 1.10, p = 0.031).","one_line_summary":"A fleet of sensor-specialized 22M-parameter JEPA models routed by an LLM improves LLM-as-judge scores on hydrologic questions over AlphaEarth alone with Cohen's d of 1.10.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the five Mini-JEPA manifolds capture hydrologic signals genuinely absent from AlphaEarth and that the router LLM's perfect hit rate on the curated question set will hold for real user queries outside the test distribution.","pith_extraction_headline":"A fleet of five small specialized foundation models routed by an LLM improves hydrologic reasoning over a single generalist model."},"references":{"count":30,"sample":[{"doi":"","year":2025,"title":"AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data","work_id":"86dc003c-e312-4f37-9b21-36673f0ffca8","ref_index":1,"cited_arxiv_id":"2507.22291","is_internal_anchor":true},{"doi":"10.1038/s41586-025-09005-y","year":2025,"title":"Bruinsma, Ana Lucic, Megan Stanley, Anna Allen, Johannes Brandstetter, Patrick Garvan, Maik Riechert, Jonathan A","work_id":"5fe15094-45e1-420c-9313-248fecbf4502","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Foundation models for generalist geospatial artificial intelligence","work_id":"8fc24702-11b6-4efd-87e6-70dabf8dd108","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"A. Xiao, W. Xuan, J. Wang, J. Huang, D. Tao, S. Lu, N. Yokoya, Foundation models for remote sensing and earth observation: A survey, IEEE Geoscience and Remote Sensing Magazine (2025)","work_id":"743237fa-0d37-4064-9958-9484a02d823b","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"On the opportunities and challenges of foundation models for geospatial artificial intelligence","work_id":"22c7ffb9-de2c-4523-a463-9c8192ca47d6","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":30,"snapshot_sha256":"54ae1634b8dd3b118b3b8c8ecbb990fb57b93c4f1841d96400c6242e3570d660","internal_anchors":1},"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":"d54fdc04-6275-4eb2-b9b5-bb375ffbf311"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"urtONvK2PP77y0XetnujzEFhgKWWok+tIOAio62Z1g/KTXqk6OfLePHjZBve3SBUApIsP92NP1UpLt6ke0I0AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-18T17:25:11.752893Z"},"content_sha256":"da99f91ca7c226d9784e100ee2cee4048c499449047b07545d6c5b7b982dd829","schema_version":"1.0","event_id":"sha256:da99f91ca7c226d9784e100ee2cee4048c499449047b07545d6c5b7b982dd829"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LGXGNG4SNSXU7ZUSQBC7S37F2D/bundle.json","state_url":"https://pith.science/pith/LGXGNG4SNSXU7ZUSQBC7S37F2D/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LGXGNG4SNSXU7ZUSQBC7S37F2D/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":"2026-05-18T17:25:11Z","links":{"resolver":"https://pith.science/pith/LGXGNG4SNSXU7ZUSQBC7S37F2D","bundle":"https://pith.science/pith/LGXGNG4SNSXU7ZUSQBC7S37F2D/bundle.json","state":"https://pith.science/pith/LGXGNG4SNSXU7ZUSQBC7S37F2D/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LGXGNG4SNSXU7ZUSQBC7S37F2D/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:LGXGNG4SNSXU7ZUSQBC7S37F2D","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":"ade8fde49fb7ece80554c13675da1ea7c1fe6bde903b0dab5674383cc26aea8b","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T21:11:14Z","title_canon_sha256":"63d3eaa1aff67688201a21e2e89160866d6a927959f158422f7598e99902f47e"},"schema_version":"1.0","source":{"id":"2605.14120","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14120","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14120v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14120","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"LGXGNG4SNSXU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"LGXGNG4SNSXU7ZUS","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"LGXGNG4S","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:da99f91ca7c226d9784e100ee2cee4048c499449047b07545d6c5b7b982dd829","target":"graph","created_at":"2026-05-17T23:39:11Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"In paired LLM-as-Judge evaluation, dual retrieval over AlphaEarth and the routed fleet outperforms AlphaEarth alone on physics-matched questions (Cohen's d = 1.10, p = 0.031)."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the five Mini-JEPA manifolds capture hydrologic signals genuinely absent from AlphaEarth and that the router LLM's perfect hit rate on the curated question set will hold for real user queries outside the test distribution."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A fleet of sensor-specialized 22M-parameter JEPA models routed by an LLM improves LLM-as-judge scores on hydrologic questions over AlphaEarth alone with Cohen's d of 1.10."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A fleet of five small specialized foundation models routed by an LLM improves hydrologic reasoning over a single generalist model."}],"snapshot_sha256":"4a99564c04f01b4a593cee2be331f9bd447afbbedc40185a5a10cf772e12db79"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Geospatial foundation models compress multispectral observations into dense embeddings increasingly used in natural-language environmental reasoning systems. A single planetary-scale model, e.g. Google AlphaEarth, handles broad characterization well but may compromise on specialized hydrologic signals. Such generalist models are also often inaccessible, expensive, and require large-scale compute. We propose Mini-JEPAs: a fleet of small sensor-specialized Joint Embedding Predictive Architecture (JEPA) foundation models consulted by a routing agent for specialized questions. We pretrained five 2","authors_text":"Mashrekur Rahman","cross_cats":["cs.CL"],"headline":"A fleet of five small specialized foundation models routed by an LLM improves hydrologic reasoning over a single generalist model.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T21:11:14Z","title":"Mini-JEPA Foundation Model Fleet Enables Agentic Hydrologic Intelligence"},"references":{"count":30,"internal_anchors":1,"resolved_work":30,"sample":[{"cited_arxiv_id":"2507.22291","doi":"","is_internal_anchor":true,"ref_index":1,"title":"AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data","work_id":"86dc003c-e312-4f37-9b21-36673f0ffca8","year":2025},{"cited_arxiv_id":"","doi":"10.1038/s41586-025-09005-y","is_internal_anchor":false,"ref_index":2,"title":"Bruinsma, Ana Lucic, Megan Stanley, Anna Allen, Johannes Brandstetter, Patrick Garvan, Maik Riechert, Jonathan A","work_id":"5fe15094-45e1-420c-9313-248fecbf4502","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Foundation models for generalist geospatial artificial intelligence","work_id":"8fc24702-11b6-4efd-87e6-70dabf8dd108","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"A. Xiao, W. Xuan, J. Wang, J. Huang, D. Tao, S. Lu, N. Yokoya, Foundation models for remote sensing and earth observation: A survey, IEEE Geoscience and Remote Sensing Magazine (2025)","work_id":"743237fa-0d37-4064-9958-9484a02d823b","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"On the opportunities and challenges of foundation models for geospatial artificial intelligence","work_id":"22c7ffb9-de2c-4523-a463-9c8192ca47d6","year":2023}],"snapshot_sha256":"54ae1634b8dd3b118b3b8c8ecbb990fb57b93c4f1841d96400c6242e3570d660"},"source":{"id":"2605.14120","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T04:57:27.489456Z","id":"d54fdc04-6275-4eb2-b9b5-bb375ffbf311","model_set":{"reader":"grok-4.3"},"one_line_summary":"A fleet of sensor-specialized 22M-parameter JEPA models routed by an LLM improves LLM-as-judge scores on hydrologic questions over AlphaEarth alone with Cohen's d of 1.10.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A fleet of five small specialized foundation models routed by an LLM improves hydrologic reasoning over a single generalist model.","strongest_claim":"In paired LLM-as-Judge evaluation, dual retrieval over AlphaEarth and the routed fleet outperforms AlphaEarth alone on physics-matched questions (Cohen's d = 1.10, p = 0.031).","weakest_assumption":"That the five Mini-JEPA manifolds capture hydrologic signals genuinely absent from AlphaEarth and that the router LLM's perfect hit rate on the curated question set will hold for real user queries outside the test distribution."}},"verdict_id":"d54fdc04-6275-4eb2-b9b5-bb375ffbf311"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b75473526b2af0c8318db66181b42f5eb4ad04d4129ca39adb6466ac3dd2c3bc","target":"record","created_at":"2026-05-17T23:39:11Z","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":"ade8fde49fb7ece80554c13675da1ea7c1fe6bde903b0dab5674383cc26aea8b","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T21:11:14Z","title_canon_sha256":"63d3eaa1aff67688201a21e2e89160866d6a927959f158422f7598e99902f47e"},"schema_version":"1.0","source":{"id":"2605.14120","kind":"arxiv","version":1}},"canonical_sha256":"59ae669b926caf4fe6928045f96fe5d0d67cb6b15e4cbc6c1077e9316f8b229d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"59ae669b926caf4fe6928045f96fe5d0d67cb6b15e4cbc6c1077e9316f8b229d","first_computed_at":"2026-05-17T23:39:11.905033Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:11.905033Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yvCIhv6VmSAVi0T4hFHyvttbPYg34eJx75D7fzSzNnmCJSshVN0S3ebLd9Ysk8BsRlXuGeoun++oLg41RZeaBQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:11.905969Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14120","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b75473526b2af0c8318db66181b42f5eb4ad04d4129ca39adb6466ac3dd2c3bc","sha256:da99f91ca7c226d9784e100ee2cee4048c499449047b07545d6c5b7b982dd829"],"state_sha256":"dcc11820bc4411c280e8e7bf22c080dc020b1b2414328b8c685d6cc87230a0d8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9oAuoMSrcGcmZ8Rem+KAsdsbKWDVlBe1ItZwxYHfME8IgeRfJ4AIejyvXygGHAlK4RSc1mcZYCBoKyYj4GTiAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-18T17:25:11.754438Z","bundle_sha256":"f72f3586ad75a9009cec923f0b063072f2dba5380435aa7369a8455a68ea9e62"}}