{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:4GLJXI5JPS4MZNURHLUHQAO3NL","short_pith_number":"pith:4GLJXI5J","schema_version":"1.0","canonical_sha256":"e1969ba3a97cb8ccb6913ae87801db6aebabe0a911aa3efe081100e8e00009a2","source":{"kind":"arxiv","id":"2606.04385","version":1},"attestation_state":"computed","paper":{"title":"Geometry-Preserving Unsupervised Alignment for Heterogeneous Foundation Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Huafeng Li, Shuwen Yu, Yi Zhao, Yonghang Tai, Zhanxuan Hu","submitted_at":"2026-06-03T03:02:09Z","abstract_excerpt":"Foundation models have driven rapid progress in computer vision, yet the two dominant paradigms, vision-language foundation models (VLMs) and vision-only foundation models (VFMs), remain only partially compatible. VLMs offer language-grounded semantic alignment but are often visually coarse, while VFMs learn discriminative perceptual geometry but lack semantic grounding. We propose GPUA (Geometry-Preserving Unsupervised Alignment), a framework that integrates the complementary strengths of VFMs and VLMs. Inspired by cross-lingual alignment, GPUA treats VFM features as a visual language and lea"},"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":"2606.04385","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-03T03:02:09Z","cross_cats_sorted":[],"title_canon_sha256":"a705bef5185fd81fd1cb1e880b90e91ddfc03fdb8840806054445e8f6d36f96f","abstract_canon_sha256":"2aa97a191efa38d07e8d44268ac9beca29171bdccd5303c86c8e265ad26bad3d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:09:06.088718Z","signature_b64":"ZGq3fvhXm3lEN4IZiAE4eGNA3oF3+LuR6UxxrMj0cB7SgIJMfQjJL18znmPbZHErd11eVj2MdLws2jkfQebvCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e1969ba3a97cb8ccb6913ae87801db6aebabe0a911aa3efe081100e8e00009a2","last_reissued_at":"2026-06-04T01:09:06.088284Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:09:06.088284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Geometry-Preserving Unsupervised Alignment for Heterogeneous Foundation Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Huafeng Li, Shuwen Yu, Yi Zhao, Yonghang Tai, Zhanxuan Hu","submitted_at":"2026-06-03T03:02:09Z","abstract_excerpt":"Foundation models have driven rapid progress in computer vision, yet the two dominant paradigms, vision-language foundation models (VLMs) and vision-only foundation models (VFMs), remain only partially compatible. VLMs offer language-grounded semantic alignment but are often visually coarse, while VFMs learn discriminative perceptual geometry but lack semantic grounding. We propose GPUA (Geometry-Preserving Unsupervised Alignment), a framework that integrates the complementary strengths of VFMs and VLMs. Inspired by cross-lingual alignment, GPUA treats VFM features as a visual language and lea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04385","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/2606.04385/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":"2606.04385","created_at":"2026-06-04T01:09:06.088345+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.04385v1","created_at":"2026-06-04T01:09:06.088345+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04385","created_at":"2026-06-04T01:09:06.088345+00:00"},{"alias_kind":"pith_short_12","alias_value":"4GLJXI5JPS4M","created_at":"2026-06-04T01:09:06.088345+00:00"},{"alias_kind":"pith_short_16","alias_value":"4GLJXI5JPS4MZNUR","created_at":"2026-06-04T01:09:06.088345+00:00"},{"alias_kind":"pith_short_8","alias_value":"4GLJXI5J","created_at":"2026-06-04T01:09:06.088345+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/4GLJXI5JPS4MZNURHLUHQAO3NL","json":"https://pith.science/pith/4GLJXI5JPS4MZNURHLUHQAO3NL.json","graph_json":"https://pith.science/api/pith-number/4GLJXI5JPS4MZNURHLUHQAO3NL/graph.json","events_json":"https://pith.science/api/pith-number/4GLJXI5JPS4MZNURHLUHQAO3NL/events.json","paper":"https://pith.science/paper/4GLJXI5J"},"agent_actions":{"view_html":"https://pith.science/pith/4GLJXI5JPS4MZNURHLUHQAO3NL","download_json":"https://pith.science/pith/4GLJXI5JPS4MZNURHLUHQAO3NL.json","view_paper":"https://pith.science/paper/4GLJXI5J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.04385&json=true","fetch_graph":"https://pith.science/api/pith-number/4GLJXI5JPS4MZNURHLUHQAO3NL/graph.json","fetch_events":"https://pith.science/api/pith-number/4GLJXI5JPS4MZNURHLUHQAO3NL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4GLJXI5JPS4MZNURHLUHQAO3NL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4GLJXI5JPS4MZNURHLUHQAO3NL/action/storage_attestation","attest_author":"https://pith.science/pith/4GLJXI5JPS4MZNURHLUHQAO3NL/action/author_attestation","sign_citation":"https://pith.science/pith/4GLJXI5JPS4MZNURHLUHQAO3NL/action/citation_signature","submit_replication":"https://pith.science/pith/4GLJXI5JPS4MZNURHLUHQAO3NL/action/replication_record"}},"created_at":"2026-06-04T01:09:06.088345+00:00","updated_at":"2026-06-04T01:09:06.088345+00:00"}