{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:K6HYSKWDOWQPCTCABEGX4IKFAS","short_pith_number":"pith:K6HYSKWD","schema_version":"1.0","canonical_sha256":"578f892ac375a0f14c40090d7e2145048a8f094fc18920dbafafbc766532f1a9","source":{"kind":"arxiv","id":"2407.05010","version":1},"attestation_state":"computed","paper":{"title":"PRANCE: Joint Token-Optimization and Structural Channel-Pruning for Adaptive ViT Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Tang, Jiajun Fan, Wenwu Zhu, Xinzhu Ma, Ye Li, Yuan Meng, Zenghao Chai, Zhi Wang","submitted_at":"2024-07-06T09:04:27Z","abstract_excerpt":"We introduce PRANCE, a Vision Transformer compression framework that jointly optimizes the activated channels and reduces tokens, based on the characteristics of inputs. Specifically, PRANCE~ leverages adaptive token optimization strategies for a certain computational budget, aiming to accelerate ViTs' inference from a unified data and architectural perspective. However, the joint framework poses challenges to both architectural and decision-making aspects. Firstly, while ViTs inherently support variable-token inference, they do not facilitate dynamic computations for variable channels. To ove"},"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":"2407.05010","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-07-06T09:04:27Z","cross_cats_sorted":[],"title_canon_sha256":"9ae0c402f27bbff11fed80c1d7d9ff5f760688f86a408a5df777268a764d40aa","abstract_canon_sha256":"e016d2e33356f2cb364aa4b225d110a02b7add03c25bcad9e40ee9c0df3ce170"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:41:06.611854Z","signature_b64":"iwmNuaaoETbcmD0AgUAifK90lsbbtENjSGdX1HjT6hOZfNHUWR5fQMahSX0olb8pafqRvLQpVBqfAfI8xYymDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"578f892ac375a0f14c40090d7e2145048a8f094fc18920dbafafbc766532f1a9","last_reissued_at":"2026-07-05T08:41:06.611364Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:41:06.611364Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PRANCE: Joint Token-Optimization and Structural Channel-Pruning for Adaptive ViT Inference","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Tang, Jiajun Fan, Wenwu Zhu, Xinzhu Ma, Ye Li, Yuan Meng, Zenghao Chai, Zhi Wang","submitted_at":"2024-07-06T09:04:27Z","abstract_excerpt":"We introduce PRANCE, a Vision Transformer compression framework that jointly optimizes the activated channels and reduces tokens, based on the characteristics of inputs. Specifically, PRANCE~ leverages adaptive token optimization strategies for a certain computational budget, aiming to accelerate ViTs' inference from a unified data and architectural perspective. However, the joint framework poses challenges to both architectural and decision-making aspects. Firstly, while ViTs inherently support variable-token inference, they do not facilitate dynamic computations for variable channels. To ove"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.05010","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/2407.05010/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":"2407.05010","created_at":"2026-07-05T08:41:06.611423+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.05010v1","created_at":"2026-07-05T08:41:06.611423+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.05010","created_at":"2026-07-05T08:41:06.611423+00:00"},{"alias_kind":"pith_short_12","alias_value":"K6HYSKWDOWQP","created_at":"2026-07-05T08:41:06.611423+00:00"},{"alias_kind":"pith_short_16","alias_value":"K6HYSKWDOWQPCTCA","created_at":"2026-07-05T08:41:06.611423+00:00"},{"alias_kind":"pith_short_8","alias_value":"K6HYSKWD","created_at":"2026-07-05T08:41:06.611423+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.26246","citing_title":"Weak-to-Strong On-Policy Distillation","ref_index":107,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS","json":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS.json","graph_json":"https://pith.science/api/pith-number/K6HYSKWDOWQPCTCABEGX4IKFAS/graph.json","events_json":"https://pith.science/api/pith-number/K6HYSKWDOWQPCTCABEGX4IKFAS/events.json","paper":"https://pith.science/paper/K6HYSKWD"},"agent_actions":{"view_html":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS","download_json":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS.json","view_paper":"https://pith.science/paper/K6HYSKWD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.05010&json=true","fetch_graph":"https://pith.science/api/pith-number/K6HYSKWDOWQPCTCABEGX4IKFAS/graph.json","fetch_events":"https://pith.science/api/pith-number/K6HYSKWDOWQPCTCABEGX4IKFAS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS/action/storage_attestation","attest_author":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS/action/author_attestation","sign_citation":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS/action/citation_signature","submit_replication":"https://pith.science/pith/K6HYSKWDOWQPCTCABEGX4IKFAS/action/replication_record"}},"created_at":"2026-07-05T08:41:06.611423+00:00","updated_at":"2026-07-05T08:41:06.611423+00:00"}