{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:5VCS7TXX4KPIROSJCMBDDLQ556","short_pith_number":"pith:5VCS7TXX","schema_version":"1.0","canonical_sha256":"ed452fcef7e29e88ba49130231ae1def806862418f8ed7918dc7e854b1248615","source":{"kind":"arxiv","id":"1804.09235","version":2},"attestation_state":"computed","paper":{"title":"On the effectiveness of task granularity for transfer learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"David Fleet, Farzaneh Mahdisoltani, Guillaume Berger, Roland Memisevic, Waseem Gharbieh","submitted_at":"2018-04-24T20:06:55Z","abstract_excerpt":"We describe a DNN for video classification and captioning, trained end-to-end, with shared features, to solve tasks at different levels of granularity, exploring the link between granularity in a source task and the quality of learned features for transfer learning. For solving the new task domain in transfer learning, we freeze the trained encoder and fine-tune a neural net on the target domain. We train on the Something-Something dataset with over 220, 000 videos, and multiple levels of target granularity, including 50 action groups, 174 fine-grained action categories and captions. Classific"},"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":"1804.09235","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-04-24T20:06:55Z","cross_cats_sorted":[],"title_canon_sha256":"aa936e020c0cdd9f39bd4c7fb19ddb998796b3c409b6289322da3df45662cfc8","abstract_canon_sha256":"2142b89302e60c4782e9b2ec3f0584ea38f827515710e385c1dfe433697ee162"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:59:39.034111Z","signature_b64":"FS8haSSLHMDyH5+MDslvrQ8GYnom8AJgWUAE9s8FObh31oHz6noXV7MYWfIrWePlCNhcCVxWApM/WqLlWmRaCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed452fcef7e29e88ba49130231ae1def806862418f8ed7918dc7e854b1248615","last_reissued_at":"2026-05-17T23:59:39.033412Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:59:39.033412Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the effectiveness of task granularity for transfer learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"David Fleet, Farzaneh Mahdisoltani, Guillaume Berger, Roland Memisevic, Waseem Gharbieh","submitted_at":"2018-04-24T20:06:55Z","abstract_excerpt":"We describe a DNN for video classification and captioning, trained end-to-end, with shared features, to solve tasks at different levels of granularity, exploring the link between granularity in a source task and the quality of learned features for transfer learning. For solving the new task domain in transfer learning, we freeze the trained encoder and fine-tune a neural net on the target domain. We train on the Something-Something dataset with over 220, 000 videos, and multiple levels of target granularity, including 50 action groups, 174 fine-grained action categories and captions. Classific"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1804.09235","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1804.09235","created_at":"2026-05-17T23:59:39.033530+00:00"},{"alias_kind":"arxiv_version","alias_value":"1804.09235v2","created_at":"2026-05-17T23:59:39.033530+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1804.09235","created_at":"2026-05-17T23:59:39.033530+00:00"},{"alias_kind":"pith_short_12","alias_value":"5VCS7TXX4KPI","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_16","alias_value":"5VCS7TXX4KPIROSJ","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_8","alias_value":"5VCS7TXX","created_at":"2026-05-18T12:32:08.215937+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2307.06942","citing_title":"InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation","ref_index":55,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20760","citing_title":"Exploring High-Order Self-Similarity for Video Understanding","ref_index":54,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556","json":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556.json","graph_json":"https://pith.science/api/pith-number/5VCS7TXX4KPIROSJCMBDDLQ556/graph.json","events_json":"https://pith.science/api/pith-number/5VCS7TXX4KPIROSJCMBDDLQ556/events.json","paper":"https://pith.science/paper/5VCS7TXX"},"agent_actions":{"view_html":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556","download_json":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556.json","view_paper":"https://pith.science/paper/5VCS7TXX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1804.09235&json=true","fetch_graph":"https://pith.science/api/pith-number/5VCS7TXX4KPIROSJCMBDDLQ556/graph.json","fetch_events":"https://pith.science/api/pith-number/5VCS7TXX4KPIROSJCMBDDLQ556/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556/action/storage_attestation","attest_author":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556/action/author_attestation","sign_citation":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556/action/citation_signature","submit_replication":"https://pith.science/pith/5VCS7TXX4KPIROSJCMBDDLQ556/action/replication_record"}},"created_at":"2026-05-17T23:59:39.033530+00:00","updated_at":"2026-05-17T23:59:39.033530+00:00"}