{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:5GNUGXKOBASOKJFHAL47QUT2ZD","short_pith_number":"pith:5GNUGXKO","schema_version":"1.0","canonical_sha256":"e99b435d4e0824e524a702f9f8527ac8e1013d554f02900aeeb5981d861f7ecd","source":{"kind":"arxiv","id":"2102.01393","version":1},"attestation_state":"computed","paper":{"title":"It's always personal: Using Early Exits for Efficient On-Device CNN Personalisation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Ilias Leontiadis, Nicholas D. Lane, Stefanos Laskaridis, Stylianos I. Venieris","submitted_at":"2021-02-02T09:10:17Z","abstract_excerpt":"On-device machine learning is becoming a reality thanks to the availability of powerful hardware and model compression techniques. Typically, these models are pretrained on large GPU clusters and have enough parameters to generalise across a wide variety of inputs. In this work, we observe that a much smaller, personalised model can be employed to fit a specific scenario, resulting in both higher accuracy and faster execution. Nevertheless, on-device training is extremely challenging, imposing excessive computational and memory requirements even for flagship smartphones. At the same time, on-d"},"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":"2102.01393","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-02-02T09:10:17Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"fd66d0233ff44f3b642e7781eb7ab791298406a34a1583e4feac600e7cf64b3f","abstract_canon_sha256":"4b2f15278d9421c44bbe6944dbd8bcaad6aa13fe9506b2ec3e91e394a6300254"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:11:30.305382Z","signature_b64":"tgEinUa6r0bwOn5YDd+2XBaUsWcJAGMHC7c7oRj4gFiyO0o+IRcDefx2G11hsJsFbnuiSTaL+B5gFNV4j6ldAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e99b435d4e0824e524a702f9f8527ac8e1013d554f02900aeeb5981d861f7ecd","last_reissued_at":"2026-07-05T02:11:30.304943Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:11:30.304943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"It's always personal: Using Early Exits for Efficient On-Device CNN Personalisation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Ilias Leontiadis, Nicholas D. Lane, Stefanos Laskaridis, Stylianos I. Venieris","submitted_at":"2021-02-02T09:10:17Z","abstract_excerpt":"On-device machine learning is becoming a reality thanks to the availability of powerful hardware and model compression techniques. Typically, these models are pretrained on large GPU clusters and have enough parameters to generalise across a wide variety of inputs. In this work, we observe that a much smaller, personalised model can be employed to fit a specific scenario, resulting in both higher accuracy and faster execution. Nevertheless, on-device training is extremely challenging, imposing excessive computational and memory requirements even for flagship smartphones. At the same time, on-d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.01393","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/2102.01393/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":"2102.01393","created_at":"2026-07-05T02:11:30.305002+00:00"},{"alias_kind":"arxiv_version","alias_value":"2102.01393v1","created_at":"2026-07-05T02:11:30.305002+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.01393","created_at":"2026-07-05T02:11:30.305002+00:00"},{"alias_kind":"pith_short_12","alias_value":"5GNUGXKOBASO","created_at":"2026-07-05T02:11:30.305002+00:00"},{"alias_kind":"pith_short_16","alias_value":"5GNUGXKOBASOKJFH","created_at":"2026-07-05T02:11:30.305002+00:00"},{"alias_kind":"pith_short_8","alias_value":"5GNUGXKO","created_at":"2026-07-05T02:11:30.305002+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/5GNUGXKOBASOKJFHAL47QUT2ZD","json":"https://pith.science/pith/5GNUGXKOBASOKJFHAL47QUT2ZD.json","graph_json":"https://pith.science/api/pith-number/5GNUGXKOBASOKJFHAL47QUT2ZD/graph.json","events_json":"https://pith.science/api/pith-number/5GNUGXKOBASOKJFHAL47QUT2ZD/events.json","paper":"https://pith.science/paper/5GNUGXKO"},"agent_actions":{"view_html":"https://pith.science/pith/5GNUGXKOBASOKJFHAL47QUT2ZD","download_json":"https://pith.science/pith/5GNUGXKOBASOKJFHAL47QUT2ZD.json","view_paper":"https://pith.science/paper/5GNUGXKO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2102.01393&json=true","fetch_graph":"https://pith.science/api/pith-number/5GNUGXKOBASOKJFHAL47QUT2ZD/graph.json","fetch_events":"https://pith.science/api/pith-number/5GNUGXKOBASOKJFHAL47QUT2ZD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5GNUGXKOBASOKJFHAL47QUT2ZD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5GNUGXKOBASOKJFHAL47QUT2ZD/action/storage_attestation","attest_author":"https://pith.science/pith/5GNUGXKOBASOKJFHAL47QUT2ZD/action/author_attestation","sign_citation":"https://pith.science/pith/5GNUGXKOBASOKJFHAL47QUT2ZD/action/citation_signature","submit_replication":"https://pith.science/pith/5GNUGXKOBASOKJFHAL47QUT2ZD/action/replication_record"}},"created_at":"2026-07-05T02:11:30.305002+00:00","updated_at":"2026-07-05T02:11:30.305002+00:00"}