{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:SOR7MYSILD52PN55JYWWKNNYT5","short_pith_number":"pith:SOR7MYSI","schema_version":"1.0","canonical_sha256":"93a3f6624858fba7b7bd4e2d6535b89f54211cda34f093c00206c9142fd510ae","source":{"kind":"arxiv","id":"2111.15668","version":1},"attestation_state":"computed","paper":{"title":"AdaViT: Adaptive Vision Transformers for Efficient Image Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bor-Chun Chen, Hengduo Li, Lingchen Meng, Ser-Nam Lim, Shiyi Lan, Yu-Gang Jiang, Zuxuan Wu","submitted_at":"2021-11-30T18:57:02Z","abstract_excerpt":"Built on top of self-attention mechanisms, vision transformers have demonstrated remarkable performance on a variety of vision tasks recently. While achieving excellent performance, they still require relatively intensive computational cost that scales up drastically as the numbers of patches, self-attention heads and transformer blocks increase. In this paper, we argue that due to the large variations among images, their need for modeling long-range dependencies between patches differ. To this end, we introduce AdaViT, an adaptive computation framework that learns to derive usage policies on "},"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":"2111.15668","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-11-30T18:57:02Z","cross_cats_sorted":[],"title_canon_sha256":"6c76d106808ca6a9c44c4dde9241dd55a785187bdb73ae4b3fdc81dbd9987dc2","abstract_canon_sha256":"313864b7a3979a5e5246a9bf0eeadb6bb64bb71be012d4ea5f440686d8861b20"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:36:28.711603Z","signature_b64":"zRf9ho2CqApvCnRBUFpLXleKjC93qmyRWFOYMHoikZhSpnN8hvqugUirHIeF16LyVDm91XyBJyzMpFXfydXpAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"93a3f6624858fba7b7bd4e2d6535b89f54211cda34f093c00206c9142fd510ae","last_reissued_at":"2026-07-05T03:36:28.711199Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:36:28.711199Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AdaViT: Adaptive Vision Transformers for Efficient Image Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bor-Chun Chen, Hengduo Li, Lingchen Meng, Ser-Nam Lim, Shiyi Lan, Yu-Gang Jiang, Zuxuan Wu","submitted_at":"2021-11-30T18:57:02Z","abstract_excerpt":"Built on top of self-attention mechanisms, vision transformers have demonstrated remarkable performance on a variety of vision tasks recently. While achieving excellent performance, they still require relatively intensive computational cost that scales up drastically as the numbers of patches, self-attention heads and transformer blocks increase. In this paper, we argue that due to the large variations among images, their need for modeling long-range dependencies between patches differ. To this end, we introduce AdaViT, an adaptive computation framework that learns to derive usage policies on "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.15668","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/2111.15668/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":"2111.15668","created_at":"2026-07-05T03:36:28.711248+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.15668v1","created_at":"2026-07-05T03:36:28.711248+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.15668","created_at":"2026-07-05T03:36:28.711248+00:00"},{"alias_kind":"pith_short_12","alias_value":"SOR7MYSILD52","created_at":"2026-07-05T03:36:28.711248+00:00"},{"alias_kind":"pith_short_16","alias_value":"SOR7MYSILD52PN55","created_at":"2026-07-05T03:36:28.711248+00:00"},{"alias_kind":"pith_short_8","alias_value":"SOR7MYSI","created_at":"2026-07-05T03:36:28.711248+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/SOR7MYSILD52PN55JYWWKNNYT5","json":"https://pith.science/pith/SOR7MYSILD52PN55JYWWKNNYT5.json","graph_json":"https://pith.science/api/pith-number/SOR7MYSILD52PN55JYWWKNNYT5/graph.json","events_json":"https://pith.science/api/pith-number/SOR7MYSILD52PN55JYWWKNNYT5/events.json","paper":"https://pith.science/paper/SOR7MYSI"},"agent_actions":{"view_html":"https://pith.science/pith/SOR7MYSILD52PN55JYWWKNNYT5","download_json":"https://pith.science/pith/SOR7MYSILD52PN55JYWWKNNYT5.json","view_paper":"https://pith.science/paper/SOR7MYSI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.15668&json=true","fetch_graph":"https://pith.science/api/pith-number/SOR7MYSILD52PN55JYWWKNNYT5/graph.json","fetch_events":"https://pith.science/api/pith-number/SOR7MYSILD52PN55JYWWKNNYT5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SOR7MYSILD52PN55JYWWKNNYT5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SOR7MYSILD52PN55JYWWKNNYT5/action/storage_attestation","attest_author":"https://pith.science/pith/SOR7MYSILD52PN55JYWWKNNYT5/action/author_attestation","sign_citation":"https://pith.science/pith/SOR7MYSILD52PN55JYWWKNNYT5/action/citation_signature","submit_replication":"https://pith.science/pith/SOR7MYSILD52PN55JYWWKNNYT5/action/replication_record"}},"created_at":"2026-07-05T03:36:28.711248+00:00","updated_at":"2026-07-05T03:36:28.711248+00:00"}