{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VSET767ZF3GQ45K4RHRIADREJU","short_pith_number":"pith:VSET767Z","schema_version":"1.0","canonical_sha256":"ac893ffbf92ecd0e755c89e2800e244d3d5a7c97bfef0b05c7dc6b0a856ad969","source":{"kind":"arxiv","id":"2505.18758","version":1},"attestation_state":"computed","paper":{"title":"Reducing Storage of Pretrained Neural Networks by Rate-Constrained Quantization and Entropy Coding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander Conzelmann, Robert Bamler","submitted_at":"2025-05-24T15:52:49Z","abstract_excerpt":"The ever-growing size of neural networks poses serious challenges on resource-constrained devices, such as embedded sensors. Compression algorithms that reduce their size can mitigate these problems, provided that model performance stays close to the original. We propose a novel post-training compression framework that combines rate-aware quantization with entropy coding by (1) extending the well-known layer-wise loss by a quadratic rate estimation, and (2) providing locally exact solutions to this modified objective following the Optimal Brain Surgeon (OBS) method. Our method allows for very "},"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":"2505.18758","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-24T15:52:49Z","cross_cats_sorted":[],"title_canon_sha256":"6c3d0f5244197afe10bbf6e80a15d464abefb2d4e4ce52e6276f1a2e9feb2511","abstract_canon_sha256":"150e21cd9313ac2b25ef6172e61cf7c2acb4c4e2cc76a3b6c46a17c78cbe777a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:09:19.165117Z","signature_b64":"MGrxrCjLBAgcBHHpSJytfIYg6ZqthT0VLz/s3ayjn90gmQcNFYS4pI5v8PpdWLrR5vSSHAKL+P4xX7j87tOpAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ac893ffbf92ecd0e755c89e2800e244d3d5a7c97bfef0b05c7dc6b0a856ad969","last_reissued_at":"2026-07-05T11:09:19.164677Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:09:19.164677Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reducing Storage of Pretrained Neural Networks by Rate-Constrained Quantization and Entropy Coding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander Conzelmann, Robert Bamler","submitted_at":"2025-05-24T15:52:49Z","abstract_excerpt":"The ever-growing size of neural networks poses serious challenges on resource-constrained devices, such as embedded sensors. Compression algorithms that reduce their size can mitigate these problems, provided that model performance stays close to the original. We propose a novel post-training compression framework that combines rate-aware quantization with entropy coding by (1) extending the well-known layer-wise loss by a quadratic rate estimation, and (2) providing locally exact solutions to this modified objective following the Optimal Brain Surgeon (OBS) method. Our method allows for very "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18758","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/2505.18758/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":"2505.18758","created_at":"2026-07-05T11:09:19.164733+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18758v1","created_at":"2026-07-05T11:09:19.164733+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18758","created_at":"2026-07-05T11:09:19.164733+00:00"},{"alias_kind":"pith_short_12","alias_value":"VSET767ZF3GQ","created_at":"2026-07-05T11:09:19.164733+00:00"},{"alias_kind":"pith_short_16","alias_value":"VSET767ZF3GQ45K4","created_at":"2026-07-05T11:09:19.164733+00:00"},{"alias_kind":"pith_short_8","alias_value":"VSET767Z","created_at":"2026-07-05T11:09:19.164733+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/VSET767ZF3GQ45K4RHRIADREJU","json":"https://pith.science/pith/VSET767ZF3GQ45K4RHRIADREJU.json","graph_json":"https://pith.science/api/pith-number/VSET767ZF3GQ45K4RHRIADREJU/graph.json","events_json":"https://pith.science/api/pith-number/VSET767ZF3GQ45K4RHRIADREJU/events.json","paper":"https://pith.science/paper/VSET767Z"},"agent_actions":{"view_html":"https://pith.science/pith/VSET767ZF3GQ45K4RHRIADREJU","download_json":"https://pith.science/pith/VSET767ZF3GQ45K4RHRIADREJU.json","view_paper":"https://pith.science/paper/VSET767Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18758&json=true","fetch_graph":"https://pith.science/api/pith-number/VSET767ZF3GQ45K4RHRIADREJU/graph.json","fetch_events":"https://pith.science/api/pith-number/VSET767ZF3GQ45K4RHRIADREJU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VSET767ZF3GQ45K4RHRIADREJU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VSET767ZF3GQ45K4RHRIADREJU/action/storage_attestation","attest_author":"https://pith.science/pith/VSET767ZF3GQ45K4RHRIADREJU/action/author_attestation","sign_citation":"https://pith.science/pith/VSET767ZF3GQ45K4RHRIADREJU/action/citation_signature","submit_replication":"https://pith.science/pith/VSET767ZF3GQ45K4RHRIADREJU/action/replication_record"}},"created_at":"2026-07-05T11:09:19.164733+00:00","updated_at":"2026-07-05T11:09:19.164733+00:00"}