{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UN6PLW6NNLKGPZH4C35FJ7F4BU","short_pith_number":"pith:UN6PLW6N","schema_version":"1.0","canonical_sha256":"a37cf5dbcd6ad467e4fc16fa54fcbc0d2c3f661e1f4114d0914a20ebd80dbd01","source":{"kind":"arxiv","id":"2506.14440","version":1},"attestation_state":"computed","paper":{"title":"Model compression using knowledge distillation with integrated gradients","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"David E. Hernandez, Jose Chang, Torbj\\\"orn E. M. Nordling","submitted_at":"2025-06-17T12:00:23Z","abstract_excerpt":"Model compression is critical for deploying deep learning models on resource-constrained devices. We introduce a novel method enhancing knowledge distillation with integrated gradients (IG) as a data augmentation strategy. Our approach overlays IG maps onto input images during training, providing student models with deeper insights into teacher models' decision-making processes. Extensive evaluation on CIFAR-10 demonstrates that our IG-augmented knowledge distillation achieves 92.6% testing accuracy with a 4.1x compression factor-a significant 1.1 percentage point improvement ($p<0.001$) over "},"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":"2506.14440","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-17T12:00:23Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"cfa6b5d183f1f3460c0a7ceaa1ede7525bb57919d0a4f871f28ed1569d29a524","abstract_canon_sha256":"0f8428659ef405ff9f9c40a95cf7fe6c0e8938e3b8901977c07eb21c5dda8deb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:56.590726Z","signature_b64":"aakKlBW8fstoN63NxSEycUdag4Rg8uMGWwwjfOw9RKy8dPPYHQIiejgHSUKurNor52d3C9XrYOqdZIh+UrrtCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a37cf5dbcd6ad467e4fc16fa54fcbc0d2c3f661e1f4114d0914a20ebd80dbd01","last_reissued_at":"2026-07-05T11:22:56.590153Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:56.590153Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Model compression using knowledge distillation with integrated gradients","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"David E. Hernandez, Jose Chang, Torbj\\\"orn E. M. Nordling","submitted_at":"2025-06-17T12:00:23Z","abstract_excerpt":"Model compression is critical for deploying deep learning models on resource-constrained devices. We introduce a novel method enhancing knowledge distillation with integrated gradients (IG) as a data augmentation strategy. Our approach overlays IG maps onto input images during training, providing student models with deeper insights into teacher models' decision-making processes. Extensive evaluation on CIFAR-10 demonstrates that our IG-augmented knowledge distillation achieves 92.6% testing accuracy with a 4.1x compression factor-a significant 1.1 percentage point improvement ($p<0.001$) over "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.14440","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/2506.14440/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":"2506.14440","created_at":"2026-07-05T11:22:56.590222+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.14440v1","created_at":"2026-07-05T11:22:56.590222+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.14440","created_at":"2026-07-05T11:22:56.590222+00:00"},{"alias_kind":"pith_short_12","alias_value":"UN6PLW6NNLKG","created_at":"2026-07-05T11:22:56.590222+00:00"},{"alias_kind":"pith_short_16","alias_value":"UN6PLW6NNLKGPZH4","created_at":"2026-07-05T11:22:56.590222+00:00"},{"alias_kind":"pith_short_8","alias_value":"UN6PLW6N","created_at":"2026-07-05T11:22:56.590222+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/UN6PLW6NNLKGPZH4C35FJ7F4BU","json":"https://pith.science/pith/UN6PLW6NNLKGPZH4C35FJ7F4BU.json","graph_json":"https://pith.science/api/pith-number/UN6PLW6NNLKGPZH4C35FJ7F4BU/graph.json","events_json":"https://pith.science/api/pith-number/UN6PLW6NNLKGPZH4C35FJ7F4BU/events.json","paper":"https://pith.science/paper/UN6PLW6N"},"agent_actions":{"view_html":"https://pith.science/pith/UN6PLW6NNLKGPZH4C35FJ7F4BU","download_json":"https://pith.science/pith/UN6PLW6NNLKGPZH4C35FJ7F4BU.json","view_paper":"https://pith.science/paper/UN6PLW6N","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.14440&json=true","fetch_graph":"https://pith.science/api/pith-number/UN6PLW6NNLKGPZH4C35FJ7F4BU/graph.json","fetch_events":"https://pith.science/api/pith-number/UN6PLW6NNLKGPZH4C35FJ7F4BU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UN6PLW6NNLKGPZH4C35FJ7F4BU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UN6PLW6NNLKGPZH4C35FJ7F4BU/action/storage_attestation","attest_author":"https://pith.science/pith/UN6PLW6NNLKGPZH4C35FJ7F4BU/action/author_attestation","sign_citation":"https://pith.science/pith/UN6PLW6NNLKGPZH4C35FJ7F4BU/action/citation_signature","submit_replication":"https://pith.science/pith/UN6PLW6NNLKGPZH4C35FJ7F4BU/action/replication_record"}},"created_at":"2026-07-05T11:22:56.590222+00:00","updated_at":"2026-07-05T11:22:56.590222+00:00"}