{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SN2RFZCKF4SPTDY4MTBR5EU72Q","short_pith_number":"pith:SN2RFZCK","schema_version":"1.0","canonical_sha256":"937512e44a2f24f98f1c64c31e929fd4307f5b73cfa212e67fdb12f247da093c","source":{"kind":"arxiv","id":"2607.22205","version":1},"attestation_state":"computed","paper":{"title":"Filling Before Advancing: Capability-Gap-Driven Post-Training for Scenario-Specialized Remote Sensing MLLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Antonio Plaza, Jon Atli Benediktsson, Minghua Wang, Xin Zhao, Yuheng Zong, Zhi-Hui Zhan","submitted_at":"2026-07-24T11:21:25Z","abstract_excerpt":"Remote sensing multimodal large language models (RS-MLLMs) have improved general aerial-image understanding. However, Earth observation applications require fine-grained scenario specialization, constrained by scarce high-quality scenario data and incomplete capability coverage. We formulate this adaptation as a capability-gap-driven post-training problem and propose filling before advancing (FBA). Rather than relying on single-stage supervised fine-tuning (SFT) over target-domain samples, FBA first fills prerequisite capability gaps before advancing toward scenario specialization. We instanti"},"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":"2607.22205","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-24T11:21:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"242c261c6fee515474bfe0a823c4263eb89e3ff8fd77383b069b731c96f993e7","abstract_canon_sha256":"2ec6cf31c5df279f6e72dd3f3ea3490b4730887fa95209a3e5120611e8679022"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-27T01:20:53.786067Z","signature_b64":"MbEOQjzIyEiPfOabiGjNLsZ93Rh+qLl0ZsxMfWujaDNnXgTer+gNWrXicpCcgDDfmfr3M1ryV7dmRjS1wQz1DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"937512e44a2f24f98f1c64c31e929fd4307f5b73cfa212e67fdb12f247da093c","last_reissued_at":"2026-07-27T01:20:53.785148Z","signature_status":"signed_v1","first_computed_at":"2026-07-27T01:20:53.785148Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Filling Before Advancing: Capability-Gap-Driven Post-Training for Scenario-Specialized Remote Sensing MLLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Antonio Plaza, Jon Atli Benediktsson, Minghua Wang, Xin Zhao, Yuheng Zong, Zhi-Hui Zhan","submitted_at":"2026-07-24T11:21:25Z","abstract_excerpt":"Remote sensing multimodal large language models (RS-MLLMs) have improved general aerial-image understanding. However, Earth observation applications require fine-grained scenario specialization, constrained by scarce high-quality scenario data and incomplete capability coverage. We formulate this adaptation as a capability-gap-driven post-training problem and propose filling before advancing (FBA). Rather than relying on single-stage supervised fine-tuning (SFT) over target-domain samples, FBA first fills prerequisite capability gaps before advancing toward scenario specialization. We instanti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22205","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/2607.22205/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":"2607.22205","created_at":"2026-07-27T01:20:53.785629+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.22205v1","created_at":"2026-07-27T01:20:53.785629+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22205","created_at":"2026-07-27T01:20:53.785629+00:00"},{"alias_kind":"pith_short_12","alias_value":"SN2RFZCKF4SP","created_at":"2026-07-27T01:20:53.785629+00:00"},{"alias_kind":"pith_short_16","alias_value":"SN2RFZCKF4SPTDY4","created_at":"2026-07-27T01:20:53.785629+00:00"},{"alias_kind":"pith_short_8","alias_value":"SN2RFZCK","created_at":"2026-07-27T01:20:53.785629+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/SN2RFZCKF4SPTDY4MTBR5EU72Q","json":"https://pith.science/pith/SN2RFZCKF4SPTDY4MTBR5EU72Q.json","graph_json":"https://pith.science/api/pith-number/SN2RFZCKF4SPTDY4MTBR5EU72Q/graph.json","events_json":"https://pith.science/api/pith-number/SN2RFZCKF4SPTDY4MTBR5EU72Q/events.json","paper":"https://pith.science/paper/SN2RFZCK"},"agent_actions":{"view_html":"https://pith.science/pith/SN2RFZCKF4SPTDY4MTBR5EU72Q","download_json":"https://pith.science/pith/SN2RFZCKF4SPTDY4MTBR5EU72Q.json","view_paper":"https://pith.science/paper/SN2RFZCK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.22205&json=true","fetch_graph":"https://pith.science/api/pith-number/SN2RFZCKF4SPTDY4MTBR5EU72Q/graph.json","fetch_events":"https://pith.science/api/pith-number/SN2RFZCKF4SPTDY4MTBR5EU72Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SN2RFZCKF4SPTDY4MTBR5EU72Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SN2RFZCKF4SPTDY4MTBR5EU72Q/action/storage_attestation","attest_author":"https://pith.science/pith/SN2RFZCKF4SPTDY4MTBR5EU72Q/action/author_attestation","sign_citation":"https://pith.science/pith/SN2RFZCKF4SPTDY4MTBR5EU72Q/action/citation_signature","submit_replication":"https://pith.science/pith/SN2RFZCKF4SPTDY4MTBR5EU72Q/action/replication_record"}},"created_at":"2026-07-27T01:20:53.785629+00:00","updated_at":"2026-07-27T01:20:53.785629+00:00"}