{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:4CGKOIB3JEKEN3QVJ6SZYOOVDL","short_pith_number":"pith:4CGKOIB3","schema_version":"1.0","canonical_sha256":"e08ca7203b491446ee154fa59c39d51ad457c138c2fe0ae748bfe36844caa0e9","source":{"kind":"arxiv","id":"2310.07265","version":1},"attestation_state":"computed","paper":{"title":"Distilling Efficient Vision Transformers from CNNs for Semantic Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lin Wang, Pengyuan Zhou, Xu Zheng, Yunhao Luo","submitted_at":"2023-10-11T07:45:37Z","abstract_excerpt":"In this paper, we tackle a new problem: how to transfer knowledge from the pre-trained cumbersome yet well-performed CNN-based model to learn a compact Vision Transformer (ViT)-based model while maintaining its learning capacity? Due to the completely different characteristics of ViT and CNN and the long-existing capacity gap between teacher and student models in Knowledge Distillation (KD), directly transferring the cross-model knowledge is non-trivial. To this end, we subtly leverage the visual and linguistic-compatible feature character of ViT (i.e., student), and its capacity gap with the "},"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":"2310.07265","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-10-11T07:45:37Z","cross_cats_sorted":[],"title_canon_sha256":"7a26c8d487ff688936e8e4c283a96a2a5763bac67f4ca46f3c737730fb82014c","abstract_canon_sha256":"22f5b6c7720e2f2bd083f0ebfb2b0eb41735c3abb1eb1b66d5da9750bf69f483"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:59:42.848200Z","signature_b64":"yRuUzbdIWlziw+NeJ5SsCfVlhJM/nfArkANVVp+K1KTHHinb6pW65FDJD4HwyVQMrmdvtOPPZSzrJdULhdAwDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e08ca7203b491446ee154fa59c39d51ad457c138c2fe0ae748bfe36844caa0e9","last_reissued_at":"2026-07-05T06:59:42.847726Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:59:42.847726Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distilling Efficient Vision Transformers from CNNs for Semantic Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lin Wang, Pengyuan Zhou, Xu Zheng, Yunhao Luo","submitted_at":"2023-10-11T07:45:37Z","abstract_excerpt":"In this paper, we tackle a new problem: how to transfer knowledge from the pre-trained cumbersome yet well-performed CNN-based model to learn a compact Vision Transformer (ViT)-based model while maintaining its learning capacity? Due to the completely different characteristics of ViT and CNN and the long-existing capacity gap between teacher and student models in Knowledge Distillation (KD), directly transferring the cross-model knowledge is non-trivial. To this end, we subtly leverage the visual and linguistic-compatible feature character of ViT (i.e., student), and its capacity gap with the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.07265","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/2310.07265/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":"2310.07265","created_at":"2026-07-05T06:59:42.847781+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.07265v1","created_at":"2026-07-05T06:59:42.847781+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.07265","created_at":"2026-07-05T06:59:42.847781+00:00"},{"alias_kind":"pith_short_12","alias_value":"4CGKOIB3JEKE","created_at":"2026-07-05T06:59:42.847781+00:00"},{"alias_kind":"pith_short_16","alias_value":"4CGKOIB3JEKEN3QV","created_at":"2026-07-05T06:59:42.847781+00:00"},{"alias_kind":"pith_short_8","alias_value":"4CGKOIB3","created_at":"2026-07-05T06:59:42.847781+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.16876","citing_title":"MAGIC++: Efficient and Resilient Modality-Agnostic Semantic Segmentation via Hierarchical Modality Selection","ref_index":23,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL","json":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL.json","graph_json":"https://pith.science/api/pith-number/4CGKOIB3JEKEN3QVJ6SZYOOVDL/graph.json","events_json":"https://pith.science/api/pith-number/4CGKOIB3JEKEN3QVJ6SZYOOVDL/events.json","paper":"https://pith.science/paper/4CGKOIB3"},"agent_actions":{"view_html":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL","download_json":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL.json","view_paper":"https://pith.science/paper/4CGKOIB3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.07265&json=true","fetch_graph":"https://pith.science/api/pith-number/4CGKOIB3JEKEN3QVJ6SZYOOVDL/graph.json","fetch_events":"https://pith.science/api/pith-number/4CGKOIB3JEKEN3QVJ6SZYOOVDL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL/action/storage_attestation","attest_author":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL/action/author_attestation","sign_citation":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL/action/citation_signature","submit_replication":"https://pith.science/pith/4CGKOIB3JEKEN3QVJ6SZYOOVDL/action/replication_record"}},"created_at":"2026-07-05T06:59:42.847781+00:00","updated_at":"2026-07-05T06:59:42.847781+00:00"}