{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ATSHIBRGMJP2CSKUGNHA733XDO","short_pith_number":"pith:ATSHIBRG","schema_version":"1.0","canonical_sha256":"04e4740626625fa14954334e0fef771bbaf7738d0e90fc6d9a339e8788abeabf","source":{"kind":"arxiv","id":"2310.11611","version":1},"attestation_state":"computed","paper":{"title":"In defense of parameter sharing for model-compression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aditya Desai, Anshumali Shrivastava","submitted_at":"2023-10-17T22:08:01Z","abstract_excerpt":"When considering a model architecture, there are several ways to reduce its memory footprint. Historically, popular approaches included selecting smaller architectures and creating sparse networks through pruning. More recently, randomized parameter-sharing (RPS) methods have gained traction for model compression at start of training. In this paper, we comprehensively assess the trade-off between memory and accuracy across RPS, pruning techniques, and building smaller models. Our findings demonstrate that RPS, which is both data and model-agnostic, consistently outperforms/matches smaller mode"},"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":"2310.11611","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-17T22:08:01Z","cross_cats_sorted":[],"title_canon_sha256":"0b7bf34c53aa5a99b70869f0da61a4519c119914ed8cbf74068f46f18489cf21","abstract_canon_sha256":"2bd4a2de2af95099eefc5053a756d6d6e8d6e2f9ce954a8dcb7e8c7df8710c48"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:02:09.906519Z","signature_b64":"2UZTQEaGPmBJKOlV7EIHX9Y0eR44ugOwB+dexIs0ar54Rkny9rw/4uB3vMzEy9ZEyNcjP/Awfwxs7AlCEVJmBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"04e4740626625fa14954334e0fef771bbaf7738d0e90fc6d9a339e8788abeabf","last_reissued_at":"2026-07-05T07:02:09.906095Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:02:09.906095Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"In defense of parameter sharing for model-compression","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aditya Desai, Anshumali Shrivastava","submitted_at":"2023-10-17T22:08:01Z","abstract_excerpt":"When considering a model architecture, there are several ways to reduce its memory footprint. Historically, popular approaches included selecting smaller architectures and creating sparse networks through pruning. More recently, randomized parameter-sharing (RPS) methods have gained traction for model compression at start of training. In this paper, we comprehensively assess the trade-off between memory and accuracy across RPS, pruning techniques, and building smaller models. Our findings demonstrate that RPS, which is both data and model-agnostic, consistently outperforms/matches smaller mode"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.11611","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/2310.11611/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":"2310.11611","created_at":"2026-07-05T07:02:09.906152+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.11611v1","created_at":"2026-07-05T07:02:09.906152+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.11611","created_at":"2026-07-05T07:02:09.906152+00:00"},{"alias_kind":"pith_short_12","alias_value":"ATSHIBRGMJP2","created_at":"2026-07-05T07:02:09.906152+00:00"},{"alias_kind":"pith_short_16","alias_value":"ATSHIBRGMJP2CSKU","created_at":"2026-07-05T07:02:09.906152+00:00"},{"alias_kind":"pith_short_8","alias_value":"ATSHIBRG","created_at":"2026-07-05T07:02:09.906152+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.21724","citing_title":"Beyond N-gram: Data-Aware X-GRAM Extraction for Efficient Embedding Parameter Scaling","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO","json":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO.json","graph_json":"https://pith.science/api/pith-number/ATSHIBRGMJP2CSKUGNHA733XDO/graph.json","events_json":"https://pith.science/api/pith-number/ATSHIBRGMJP2CSKUGNHA733XDO/events.json","paper":"https://pith.science/paper/ATSHIBRG"},"agent_actions":{"view_html":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO","download_json":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO.json","view_paper":"https://pith.science/paper/ATSHIBRG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.11611&json=true","fetch_graph":"https://pith.science/api/pith-number/ATSHIBRGMJP2CSKUGNHA733XDO/graph.json","fetch_events":"https://pith.science/api/pith-number/ATSHIBRGMJP2CSKUGNHA733XDO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO/action/storage_attestation","attest_author":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO/action/author_attestation","sign_citation":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO/action/citation_signature","submit_replication":"https://pith.science/pith/ATSHIBRGMJP2CSKUGNHA733XDO/action/replication_record"}},"created_at":"2026-07-05T07:02:09.906152+00:00","updated_at":"2026-07-05T07:02:09.906152+00:00"}