{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZVAJEP5YITCF7P7Z55AGGJHGDA","short_pith_number":"pith:ZVAJEP5Y","schema_version":"1.0","canonical_sha256":"cd40923fb844c45fbff9ef406324e6180084cf9c9d52cc0f2dbc65d5c9d35c9d","source":{"kind":"arxiv","id":"2412.07679","version":2},"attestation_state":"computed","paper":{"title":"RADIOv2.5: Improved Baselines for Agglomerative Vision Foundation Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Andrew Tao, Bryan Catanzaro, Greg Heinrich, Hongxu (Danny) Yin, Jan Kautz, Mike Ranzinger, Pavlo Molchanov, Yao Lu","submitted_at":"2024-12-10T17:06:41Z","abstract_excerpt":"Agglomerative models have recently emerged as a powerful approach to training vision foundation models, leveraging multi-teacher distillation from existing models such as CLIP, DINO, and SAM. This strategy enables the efficient creation of robust models, combining the strengths of individual teachers while significantly reducing computational and resource demands. In this paper, we thoroughly analyze state-of-the-art agglomerative models, identifying critical challenges including resolution mode shifts, teacher imbalance, idiosyncratic teacher artifacts, and an excessive number of output token"},"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":"2412.07679","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-10T17:06:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"06eab4136deae4f939f94bdba2bd0cbc35dcf806b279dd9f5a10d6fedcd782f7","abstract_canon_sha256":"1c6b11d9139699a3df40b0734ece636f28a80e2939e0693c3a2f43c0d7e0a076"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:11:29.963946Z","signature_b64":"gKa08eKrzTXeI2fzlA/UQoUrerC7OEnj0L9kKdNmjRcx9hbdzFTgxRVRs/D/I8xnEyaxBNbu/zeujtHF9YZPDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cd40923fb844c45fbff9ef406324e6180084cf9c9d52cc0f2dbc65d5c9d35c9d","last_reissued_at":"2026-07-05T10:11:29.963436Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:11:29.963436Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RADIOv2.5: Improved Baselines for Agglomerative Vision Foundation Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Andrew Tao, Bryan Catanzaro, Greg Heinrich, Hongxu (Danny) Yin, Jan Kautz, Mike Ranzinger, Pavlo Molchanov, Yao Lu","submitted_at":"2024-12-10T17:06:41Z","abstract_excerpt":"Agglomerative models have recently emerged as a powerful approach to training vision foundation models, leveraging multi-teacher distillation from existing models such as CLIP, DINO, and SAM. This strategy enables the efficient creation of robust models, combining the strengths of individual teachers while significantly reducing computational and resource demands. In this paper, we thoroughly analyze state-of-the-art agglomerative models, identifying critical challenges including resolution mode shifts, teacher imbalance, idiosyncratic teacher artifacts, and an excessive number of output token"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.07679","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/2412.07679/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":"2412.07679","created_at":"2026-07-05T10:11:29.963506+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.07679v2","created_at":"2026-07-05T10:11:29.963506+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.07679","created_at":"2026-07-05T10:11:29.963506+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZVAJEP5YITCF","created_at":"2026-07-05T10:11:29.963506+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZVAJEP5YITCF7P7Z","created_at":"2026-07-05T10:11:29.963506+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZVAJEP5Y","created_at":"2026-07-05T10:11:29.963506+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03879","citing_title":"Beyond Encoder Accumulation: Measuring Encoder Roles in Multi-Encoder VLMs","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2506.09082","citing_title":"AVA-Bench: Atomic Visual Ability Benchmark for Vision Foundation Models","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03669","citing_title":"FUS3DMaps: Scalable and Accurate Open-Vocabulary Semantic Mapping by 3D Fusion of Voxel- and Instance-Level Layers","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA","json":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA.json","graph_json":"https://pith.science/api/pith-number/ZVAJEP5YITCF7P7Z55AGGJHGDA/graph.json","events_json":"https://pith.science/api/pith-number/ZVAJEP5YITCF7P7Z55AGGJHGDA/events.json","paper":"https://pith.science/paper/ZVAJEP5Y"},"agent_actions":{"view_html":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA","download_json":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA.json","view_paper":"https://pith.science/paper/ZVAJEP5Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.07679&json=true","fetch_graph":"https://pith.science/api/pith-number/ZVAJEP5YITCF7P7Z55AGGJHGDA/graph.json","fetch_events":"https://pith.science/api/pith-number/ZVAJEP5YITCF7P7Z55AGGJHGDA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA/action/storage_attestation","attest_author":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA/action/author_attestation","sign_citation":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA/action/citation_signature","submit_replication":"https://pith.science/pith/ZVAJEP5YITCF7P7Z55AGGJHGDA/action/replication_record"}},"created_at":"2026-07-05T10:11:29.963506+00:00","updated_at":"2026-07-05T10:11:29.963506+00:00"}