{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:ZYNI5JJW2K4R5XYGPEO7YEWBAN","short_pith_number":"pith:ZYNI5JJW","schema_version":"1.0","canonical_sha256":"ce1a8ea536d2b91edf06791dfc12c1037a6b459dfcf77904584f42c86b21d3fa","source":{"kind":"arxiv","id":"2206.00820","version":2},"attestation_state":"computed","paper":{"title":"NIPQ: Noise proxy-based Integrated Pseudo-Quantization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Eunhyeok Park, Juncheol Shin, Junhyuk So, Sein Park, Seungyeop Kang, Sungjoo Yoo","submitted_at":"2022-06-02T01:17:40Z","abstract_excerpt":"Straight-through estimator (STE), which enables the gradient flow over the non-differentiable function via approximation, has been favored in studies related to quantization-aware training (QAT). However, STE incurs unstable convergence during QAT, resulting in notable quality degradation in low precision. Recently, pseudoquantization training has been proposed as an alternative approach to updating the learnable parameters using the pseudo-quantization noise instead of STE. In this study, we propose a novel noise proxy-based integrated pseudoquantization (NIPQ) that enables unified support of"},"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":"2206.00820","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-06-02T01:17:40Z","cross_cats_sorted":[],"title_canon_sha256":"e92afa15bb9b72fa2cb9a5037f3a902d2e6d1d9266640c1900d4e8b6b18de673","abstract_canon_sha256":"7723bc12c46436cc092c67b7dfc4f36f44206717e03edf6c5e1d6e094c5cc0dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:26:40.738902Z","signature_b64":"8M3BjN3jAbCzXZKjaAdevUlFfOR/g51K+HKdnEC/CC7vRuOB8UgbumIWHycTZSIoxMRy2g+yj93ptcwRqV2hCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce1a8ea536d2b91edf06791dfc12c1037a6b459dfcf77904584f42c86b21d3fa","last_reissued_at":"2026-07-05T06:26:40.738354Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:26:40.738354Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NIPQ: Noise proxy-based Integrated Pseudo-Quantization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Eunhyeok Park, Juncheol Shin, Junhyuk So, Sein Park, Seungyeop Kang, Sungjoo Yoo","submitted_at":"2022-06-02T01:17:40Z","abstract_excerpt":"Straight-through estimator (STE), which enables the gradient flow over the non-differentiable function via approximation, has been favored in studies related to quantization-aware training (QAT). However, STE incurs unstable convergence during QAT, resulting in notable quality degradation in low precision. Recently, pseudoquantization training has been proposed as an alternative approach to updating the learnable parameters using the pseudo-quantization noise instead of STE. In this study, we propose a novel noise proxy-based integrated pseudoquantization (NIPQ) that enables unified support of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.00820","kind":"arxiv","version":2},"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/2206.00820/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":"2206.00820","created_at":"2026-07-05T06:26:40.738411+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.00820v2","created_at":"2026-07-05T06:26:40.738411+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.00820","created_at":"2026-07-05T06:26:40.738411+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZYNI5JJW2K4R","created_at":"2026-07-05T06:26:40.738411+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZYNI5JJW2K4R5XYG","created_at":"2026-07-05T06:26:40.738411+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZYNI5JJW","created_at":"2026-07-05T06:26:40.738411+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.11170","citing_title":"Gaussian Weight Sampling for Scalable, Efficient and Stable Pseudo-Quantization Training","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN","json":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN.json","graph_json":"https://pith.science/api/pith-number/ZYNI5JJW2K4R5XYGPEO7YEWBAN/graph.json","events_json":"https://pith.science/api/pith-number/ZYNI5JJW2K4R5XYGPEO7YEWBAN/events.json","paper":"https://pith.science/paper/ZYNI5JJW"},"agent_actions":{"view_html":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN","download_json":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN.json","view_paper":"https://pith.science/paper/ZYNI5JJW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.00820&json=true","fetch_graph":"https://pith.science/api/pith-number/ZYNI5JJW2K4R5XYGPEO7YEWBAN/graph.json","fetch_events":"https://pith.science/api/pith-number/ZYNI5JJW2K4R5XYGPEO7YEWBAN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN/action/storage_attestation","attest_author":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN/action/author_attestation","sign_citation":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN/action/citation_signature","submit_replication":"https://pith.science/pith/ZYNI5JJW2K4R5XYGPEO7YEWBAN/action/replication_record"}},"created_at":"2026-07-05T06:26:40.738411+00:00","updated_at":"2026-07-05T06:26:40.738411+00:00"}