{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:IQGO64UQFB4GBW2NCOSR4VVWLE","short_pith_number":"pith:IQGO64UQ","canonical_record":{"source":{"id":"2605.22132","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-21T08:07:23Z","cross_cats_sorted":[],"title_canon_sha256":"ebf49f7493b777f1a782a8bf00bf3474478489e6110e8ee1191b0aa7568952fa","abstract_canon_sha256":"6459432e413d6493c64d68d730c6fa5dde0346fc6fba4cbe7001fc9bbaac030c"},"schema_version":"1.0"},"canonical_sha256":"440cef7290287860db4d13a51e56b65938a69ca578a33177259a65464e164923","source":{"kind":"arxiv","id":"2605.22132","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.22132","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"arxiv_version","alias_value":"2605.22132v1","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22132","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"pith_short_12","alias_value":"IQGO64UQFB4G","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"pith_short_16","alias_value":"IQGO64UQFB4GBW2N","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"pith_short_8","alias_value":"IQGO64UQ","created_at":"2026-05-22T01:04:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:IQGO64UQFB4GBW2NCOSR4VVWLE","target":"record","payload":{"canonical_record":{"source":{"id":"2605.22132","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-21T08:07:23Z","cross_cats_sorted":[],"title_canon_sha256":"ebf49f7493b777f1a782a8bf00bf3474478489e6110e8ee1191b0aa7568952fa","abstract_canon_sha256":"6459432e413d6493c64d68d730c6fa5dde0346fc6fba4cbe7001fc9bbaac030c"},"schema_version":"1.0"},"canonical_sha256":"440cef7290287860db4d13a51e56b65938a69ca578a33177259a65464e164923","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:27.466674Z","signature_b64":"VAVz/5oqGT6bJ0gKHNLDXJO+tF+HzEapMXn2uWjs9xgyyS1cCE4/VYeoAz0FOwxuI/hsmx4odI5eit7/jJKBCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"440cef7290287860db4d13a51e56b65938a69ca578a33177259a65464e164923","last_reissued_at":"2026-05-22T01:04:27.465949Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:27.465949Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.22132","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:04:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zo8ECu6nNr4ncAoux0je1I2GuhN6Q4HgJwc6a+dcG9Y1E3VQY9CDFpy9d6vX7Ra6nROdldewGpUJyBc+8pG1BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T21:25:16.823474Z"},"content_sha256":"330efd76d7acfaacf521d6aab0df73ac42c3564a94235f57084fd7ebd2682820","schema_version":"1.0","event_id":"sha256:330efd76d7acfaacf521d6aab0df73ac42c3564a94235f57084fd7ebd2682820"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:IQGO64UQFB4GBW2NCOSR4VVWLE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Accelerating Vision Foundation Models with Drop-in Depthwise Convolution","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Carmelo Scribano, Danda Pani Paudel, Giorgia Franchini, Luc Van Gool, Marko Bertogna, Mohammad Mahdi, Nedyalko Prisadnikov, Yuqian Fu","submitted_at":"2026-05-21T08:07:23Z","abstract_excerpt":"Pretrained vision foundation models deliver strong performance across tasks with limited fine-tuning. However, their Vision Transformer (ViT) backbones impose high inference costs, limiting deployment on resource-constrained devices. In this work, we accelerate large-scale pretrained ViTs while preserving their feature extraction capabilities by exploiting the intrinsic convolution-like behavior of some attention heads. Specifically, we introduce an efficient depthwise convolution-based layer that serves as a drop-in replacement for these heads. Additionally, we propose simple strategies to id"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22132","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/2605.22132/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:04:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VZHCyZum9sFFjK/0bGArApEmErurBi5viqETGomQG2oUi4hAbe3RtgcZc389vbIl0faIdprZYqdMp8hIfhacBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T21:25:16.824150Z"},"content_sha256":"29bb12565e3b3d301cacf79162bb6de3d92df7e4fa6f1bb116bdd1ce1b32ac04","schema_version":"1.0","event_id":"sha256:29bb12565e3b3d301cacf79162bb6de3d92df7e4fa6f1bb116bdd1ce1b32ac04"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IQGO64UQFB4GBW2NCOSR4VVWLE/bundle.json","state_url":"https://pith.science/pith/IQGO64UQFB4GBW2NCOSR4VVWLE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IQGO64UQFB4GBW2NCOSR4VVWLE/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-25T21:25:16Z","links":{"resolver":"https://pith.science/pith/IQGO64UQFB4GBW2NCOSR4VVWLE","bundle":"https://pith.science/pith/IQGO64UQFB4GBW2NCOSR4VVWLE/bundle.json","state":"https://pith.science/pith/IQGO64UQFB4GBW2NCOSR4VVWLE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IQGO64UQFB4GBW2NCOSR4VVWLE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:IQGO64UQFB4GBW2NCOSR4VVWLE","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"6459432e413d6493c64d68d730c6fa5dde0346fc6fba4cbe7001fc9bbaac030c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-21T08:07:23Z","title_canon_sha256":"ebf49f7493b777f1a782a8bf00bf3474478489e6110e8ee1191b0aa7568952fa"},"schema_version":"1.0","source":{"id":"2605.22132","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.22132","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"arxiv_version","alias_value":"2605.22132v1","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22132","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"pith_short_12","alias_value":"IQGO64UQFB4G","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"pith_short_16","alias_value":"IQGO64UQFB4GBW2N","created_at":"2026-05-22T01:04:27Z"},{"alias_kind":"pith_short_8","alias_value":"IQGO64UQ","created_at":"2026-05-22T01:04:27Z"}],"graph_snapshots":[{"event_id":"sha256:29bb12565e3b3d301cacf79162bb6de3d92df7e4fa6f1bb116bdd1ce1b32ac04","target":"graph","created_at":"2026-05-22T01:04:27Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.22132/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pretrained vision foundation models deliver strong performance across tasks with limited fine-tuning. However, their Vision Transformer (ViT) backbones impose high inference costs, limiting deployment on resource-constrained devices. In this work, we accelerate large-scale pretrained ViTs while preserving their feature extraction capabilities by exploiting the intrinsic convolution-like behavior of some attention heads. Specifically, we introduce an efficient depthwise convolution-based layer that serves as a drop-in replacement for these heads. Additionally, we propose simple strategies to id","authors_text":"Carmelo Scribano, Danda Pani Paudel, Giorgia Franchini, Luc Van Gool, Marko Bertogna, Mohammad Mahdi, Nedyalko Prisadnikov, Yuqian Fu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-21T08:07:23Z","title":"Accelerating Vision Foundation Models with Drop-in Depthwise Convolution"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22132","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:330efd76d7acfaacf521d6aab0df73ac42c3564a94235f57084fd7ebd2682820","target":"record","created_at":"2026-05-22T01:04:27Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"6459432e413d6493c64d68d730c6fa5dde0346fc6fba4cbe7001fc9bbaac030c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-21T08:07:23Z","title_canon_sha256":"ebf49f7493b777f1a782a8bf00bf3474478489e6110e8ee1191b0aa7568952fa"},"schema_version":"1.0","source":{"id":"2605.22132","kind":"arxiv","version":1}},"canonical_sha256":"440cef7290287860db4d13a51e56b65938a69ca578a33177259a65464e164923","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"440cef7290287860db4d13a51e56b65938a69ca578a33177259a65464e164923","first_computed_at":"2026-05-22T01:04:27.465949Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:04:27.465949Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VAVz/5oqGT6bJ0gKHNLDXJO+tF+HzEapMXn2uWjs9xgyyS1cCE4/VYeoAz0FOwxuI/hsmx4odI5eit7/jJKBCA==","signature_status":"signed_v1","signed_at":"2026-05-22T01:04:27.466674Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.22132","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:330efd76d7acfaacf521d6aab0df73ac42c3564a94235f57084fd7ebd2682820","sha256:29bb12565e3b3d301cacf79162bb6de3d92df7e4fa6f1bb116bdd1ce1b32ac04"],"state_sha256":"fa3d1ee11d930921b3fc1f476a7d44a5f9341befa238947510b17381bcbeecf6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jSVQer4ZUseO0iCdpr1AGk93Frjoxkui8bBWkyAsLvBmVphJFw7d9R24Gxk27/ifoQRW98FEgId/GZM8cVnECA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T21:25:16.827502Z","bundle_sha256":"976ef345400f779d01a4327bb046aec9b8f9bc22296d3baf0f32a177c01dae21"}}