{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:LHZ56XF4YXKAFMXZFZ526KT2A3","short_pith_number":"pith:LHZ56XF4","schema_version":"1.0","canonical_sha256":"59f3df5cbcc5d402b2f92e7baf2a7a06ff71846fc2280dc102e7987bfb474497","source":{"kind":"arxiv","id":"2210.17409","version":2},"attestation_state":"computed","paper":{"title":"Deep Model Reassembly","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Daquan Zhou, Jingwen Ye, Songhua Liu, Xinchao Wang, Xingyi Yang","submitted_at":"2022-10-24T10:16:13Z","abstract_excerpt":"In this paper, we explore a novel knowledge-transfer task, termed as Deep Model Reassembly (DeRy), for general-purpose model reuse. Given a collection of heterogeneous models pre-trained from distinct sources and with diverse architectures, the goal of DeRy, as its name implies, is to first dissect each model into distinctive building blocks, and then selectively reassemble the derived blocks to produce customized networks under both the hardware resource and performance constraints. Such ambitious nature of DeRy inevitably imposes significant challenges, including, in the first place, the fea"},"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":"2210.17409","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2022-10-24T10:16:13Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b1fc0a295152a6ff2ac55717637dd52c9e12842ba92b1e42b1e0816a8ab613b2","abstract_canon_sha256":"03d849a78b19a8480870c56e35926f5a7d9dae0d46c2e5d0d991568563f0b017"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:44.813982Z","signature_b64":"WeiYaGwX7WcUYXeJQPNC8mUF3XAYiCNs6qZybl6A0gt/cKbNmMlq8s4xun+EqpOARPQyK98seuoSwft1/17NCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59f3df5cbcc5d402b2f92e7baf2a7a06ff71846fc2280dc102e7987bfb474497","last_reissued_at":"2026-07-05T05:12:44.813423Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:44.813423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Model Reassembly","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Daquan Zhou, Jingwen Ye, Songhua Liu, Xinchao Wang, Xingyi Yang","submitted_at":"2022-10-24T10:16:13Z","abstract_excerpt":"In this paper, we explore a novel knowledge-transfer task, termed as Deep Model Reassembly (DeRy), for general-purpose model reuse. Given a collection of heterogeneous models pre-trained from distinct sources and with diverse architectures, the goal of DeRy, as its name implies, is to first dissect each model into distinctive building blocks, and then selectively reassemble the derived blocks to produce customized networks under both the hardware resource and performance constraints. Such ambitious nature of DeRy inevitably imposes significant challenges, including, in the first place, the fea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.17409","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/2210.17409/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":"2210.17409","created_at":"2026-07-05T05:12:44.813485+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.17409v2","created_at":"2026-07-05T05:12:44.813485+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.17409","created_at":"2026-07-05T05:12:44.813485+00:00"},{"alias_kind":"pith_short_12","alias_value":"LHZ56XF4YXKA","created_at":"2026-07-05T05:12:44.813485+00:00"},{"alias_kind":"pith_short_16","alias_value":"LHZ56XF4YXKAFMXZ","created_at":"2026-07-05T05:12:44.813485+00:00"},{"alias_kind":"pith_short_8","alias_value":"LHZ56XF4","created_at":"2026-07-05T05:12:44.813485+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.00527","citing_title":"Learning Yourself: Class-Incremental Semantic Segmentation with Language-Inspired Bootstrapped Disentanglement","ref_index":50,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3","json":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3.json","graph_json":"https://pith.science/api/pith-number/LHZ56XF4YXKAFMXZFZ526KT2A3/graph.json","events_json":"https://pith.science/api/pith-number/LHZ56XF4YXKAFMXZFZ526KT2A3/events.json","paper":"https://pith.science/paper/LHZ56XF4"},"agent_actions":{"view_html":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3","download_json":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3.json","view_paper":"https://pith.science/paper/LHZ56XF4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.17409&json=true","fetch_graph":"https://pith.science/api/pith-number/LHZ56XF4YXKAFMXZFZ526KT2A3/graph.json","fetch_events":"https://pith.science/api/pith-number/LHZ56XF4YXKAFMXZFZ526KT2A3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3/action/storage_attestation","attest_author":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3/action/author_attestation","sign_citation":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3/action/citation_signature","submit_replication":"https://pith.science/pith/LHZ56XF4YXKAFMXZFZ526KT2A3/action/replication_record"}},"created_at":"2026-07-05T05:12:44.813485+00:00","updated_at":"2026-07-05T05:12:44.813485+00:00"}