{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:PZG7FRYS7YC5FMWHDQTMG4L4FO","short_pith_number":"pith:PZG7FRYS","schema_version":"1.0","canonical_sha256":"7e4df2c712fe05d2b2c71c26c3717c2b9fdf1d2ff12393a062e0c9931facc36f","source":{"kind":"arxiv","id":"1908.00821","version":1},"attestation_state":"computed","paper":{"title":"Learning Lightweight Lane Detection CNNs by Self Attention Distillation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Change Loy, Chunxiao Liu, Yuenan Hou, Zheng Ma","submitted_at":"2019-08-02T12:13:34Z","abstract_excerpt":"Training deep models for lane detection is challenging due to the very subtle and sparse supervisory signals inherent in lane annotations. Without learning from much richer context, these models often fail in challenging scenarios, e.g., severe occlusion, ambiguous lanes, and poor lighting conditions. In this paper, we present a novel knowledge distillation approach, i.e., Self Attention Distillation (SAD), which allows a model to learn from itself and gains substantial improvement without any additional supervision or labels. Specifically, we observe that attention maps extracted from a model"},"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":"1908.00821","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2019-08-02T12:13:34Z","cross_cats_sorted":[],"title_canon_sha256":"0279f43d116ff80a5600a2a227aa213eaa4009651a999fe14bd9724f6a204944","abstract_canon_sha256":"10c9b5843f2d33850713b3b6896958faa5a479d5f32060e3f086a6d0bbe48d01"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:51:09.806960Z","signature_b64":"pf7IeepLiWwst/19QA/TK4WB29/Vn/HDJk7lluJVfTudsGZeJg6cmaKsOU/9z7/nDEykIMdBbLxbtu3Tmf2xBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e4df2c712fe05d2b2c71c26c3717c2b9fdf1d2ff12393a062e0c9931facc36f","last_reissued_at":"2026-07-04T23:51:09.806503Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:51:09.806503Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Lightweight Lane Detection CNNs by Self Attention Distillation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Change Loy, Chunxiao Liu, Yuenan Hou, Zheng Ma","submitted_at":"2019-08-02T12:13:34Z","abstract_excerpt":"Training deep models for lane detection is challenging due to the very subtle and sparse supervisory signals inherent in lane annotations. Without learning from much richer context, these models often fail in challenging scenarios, e.g., severe occlusion, ambiguous lanes, and poor lighting conditions. In this paper, we present a novel knowledge distillation approach, i.e., Self Attention Distillation (SAD), which allows a model to learn from itself and gains substantial improvement without any additional supervision or labels. Specifically, we observe that attention maps extracted from a model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.00821","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/1908.00821/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":"1908.00821","created_at":"2026-07-04T23:51:09.806561+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.00821v1","created_at":"2026-07-04T23:51:09.806561+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.00821","created_at":"2026-07-04T23:51:09.806561+00:00"},{"alias_kind":"pith_short_12","alias_value":"PZG7FRYS7YC5","created_at":"2026-07-04T23:51:09.806561+00:00"},{"alias_kind":"pith_short_16","alias_value":"PZG7FRYS7YC5FMWH","created_at":"2026-07-04T23:51:09.806561+00:00"},{"alias_kind":"pith_short_8","alias_value":"PZG7FRYS","created_at":"2026-07-04T23:51:09.806561+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/PZG7FRYS7YC5FMWHDQTMG4L4FO","json":"https://pith.science/pith/PZG7FRYS7YC5FMWHDQTMG4L4FO.json","graph_json":"https://pith.science/api/pith-number/PZG7FRYS7YC5FMWHDQTMG4L4FO/graph.json","events_json":"https://pith.science/api/pith-number/PZG7FRYS7YC5FMWHDQTMG4L4FO/events.json","paper":"https://pith.science/paper/PZG7FRYS"},"agent_actions":{"view_html":"https://pith.science/pith/PZG7FRYS7YC5FMWHDQTMG4L4FO","download_json":"https://pith.science/pith/PZG7FRYS7YC5FMWHDQTMG4L4FO.json","view_paper":"https://pith.science/paper/PZG7FRYS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.00821&json=true","fetch_graph":"https://pith.science/api/pith-number/PZG7FRYS7YC5FMWHDQTMG4L4FO/graph.json","fetch_events":"https://pith.science/api/pith-number/PZG7FRYS7YC5FMWHDQTMG4L4FO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PZG7FRYS7YC5FMWHDQTMG4L4FO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PZG7FRYS7YC5FMWHDQTMG4L4FO/action/storage_attestation","attest_author":"https://pith.science/pith/PZG7FRYS7YC5FMWHDQTMG4L4FO/action/author_attestation","sign_citation":"https://pith.science/pith/PZG7FRYS7YC5FMWHDQTMG4L4FO/action/citation_signature","submit_replication":"https://pith.science/pith/PZG7FRYS7YC5FMWHDQTMG4L4FO/action/replication_record"}},"created_at":"2026-07-04T23:51:09.806561+00:00","updated_at":"2026-07-04T23:51:09.806561+00:00"}