{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3KIKEXQLWYYOP2UWMES5XUDWZ5","short_pith_number":"pith:3KIKEXQL","schema_version":"1.0","canonical_sha256":"da90a25e0bb630e7ea966125dbd076cf52f35b918afbacefda1d23b8d079e198","source":{"kind":"arxiv","id":"2606.01402","version":1},"attestation_state":"computed","paper":{"title":"Neural Network Compression by Approximate Differential Equivalence","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Andrea Passarella, Lorenzo Valerio, Mirco Tribastone, Ravi Dhiman","submitted_at":"2026-05-31T19:01:46Z","abstract_excerpt":"Neural network compression is commonly achieved by pruning parameters based on local importance scores, e.g., magnitude-based pruning. We propose a complementary approach that compresses models by aggregating neurons with similar functional behavior rather than removing weights independently. Our method encodes a trained network as a polynomial ODE system and applies a lumping method called Approximate Forward Differential Equivalence to identify neurons with approximately matching induced dynamics. A single tolerance parameter, $\\varepsilon$, controls the compression level and induces a smoot"},"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":"2606.01402","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-31T19:01:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"80309cec8bc24f58fb403896499c39ffc3a4deae5da385fe4c2067969bf482ab","abstract_canon_sha256":"ffd29d1c71dcafa9607af3f61051fdbec919d45a1ec3ec54d04eb80b24614767"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:32.422558Z","signature_b64":"qAy/lM0gYDTI/2SBPqKNFNdNvB9LfG3lKMzaZFNRDVqcrFdv0r78JovPxSinBLBQR2qb/sD7yK9fwc52NOLkCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"da90a25e0bb630e7ea966125dbd076cf52f35b918afbacefda1d23b8d079e198","last_reissued_at":"2026-06-02T02:04:32.422172Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:32.422172Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Network Compression by Approximate Differential Equivalence","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Andrea Passarella, Lorenzo Valerio, Mirco Tribastone, Ravi Dhiman","submitted_at":"2026-05-31T19:01:46Z","abstract_excerpt":"Neural network compression is commonly achieved by pruning parameters based on local importance scores, e.g., magnitude-based pruning. We propose a complementary approach that compresses models by aggregating neurons with similar functional behavior rather than removing weights independently. Our method encodes a trained network as a polynomial ODE system and applies a lumping method called Approximate Forward Differential Equivalence to identify neurons with approximately matching induced dynamics. A single tolerance parameter, $\\varepsilon$, controls the compression level and induces a smoot"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.01402","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/2606.01402/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":"2606.01402","created_at":"2026-06-02T02:04:32.422243+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.01402v1","created_at":"2026-06-02T02:04:32.422243+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.01402","created_at":"2026-06-02T02:04:32.422243+00:00"},{"alias_kind":"pith_short_12","alias_value":"3KIKEXQLWYYO","created_at":"2026-06-02T02:04:32.422243+00:00"},{"alias_kind":"pith_short_16","alias_value":"3KIKEXQLWYYOP2UW","created_at":"2026-06-02T02:04:32.422243+00:00"},{"alias_kind":"pith_short_8","alias_value":"3KIKEXQL","created_at":"2026-06-02T02:04:32.422243+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/3KIKEXQLWYYOP2UWMES5XUDWZ5","json":"https://pith.science/pith/3KIKEXQLWYYOP2UWMES5XUDWZ5.json","graph_json":"https://pith.science/api/pith-number/3KIKEXQLWYYOP2UWMES5XUDWZ5/graph.json","events_json":"https://pith.science/api/pith-number/3KIKEXQLWYYOP2UWMES5XUDWZ5/events.json","paper":"https://pith.science/paper/3KIKEXQL"},"agent_actions":{"view_html":"https://pith.science/pith/3KIKEXQLWYYOP2UWMES5XUDWZ5","download_json":"https://pith.science/pith/3KIKEXQLWYYOP2UWMES5XUDWZ5.json","view_paper":"https://pith.science/paper/3KIKEXQL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.01402&json=true","fetch_graph":"https://pith.science/api/pith-number/3KIKEXQLWYYOP2UWMES5XUDWZ5/graph.json","fetch_events":"https://pith.science/api/pith-number/3KIKEXQLWYYOP2UWMES5XUDWZ5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3KIKEXQLWYYOP2UWMES5XUDWZ5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3KIKEXQLWYYOP2UWMES5XUDWZ5/action/storage_attestation","attest_author":"https://pith.science/pith/3KIKEXQLWYYOP2UWMES5XUDWZ5/action/author_attestation","sign_citation":"https://pith.science/pith/3KIKEXQLWYYOP2UWMES5XUDWZ5/action/citation_signature","submit_replication":"https://pith.science/pith/3KIKEXQLWYYOP2UWMES5XUDWZ5/action/replication_record"}},"created_at":"2026-06-02T02:04:32.422243+00:00","updated_at":"2026-06-02T02:04:32.422243+00:00"}