{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:SZ4EVOWBXIBNXJYZ76DG37TEIM","short_pith_number":"pith:SZ4EVOWB","schema_version":"1.0","canonical_sha256":"96784abac1ba02dba719ff866dfe644319d825135930296c99cd21cd41604653","source":{"kind":"arxiv","id":"2204.04220","version":1},"attestation_state":"computed","paper":{"title":"Characterizing and Understanding the Behavior of Quantized Models for Reliable Deployment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.LG","authors_text":"Maxime Cordy, Mike Papadakis, Qiang Hu, Wei Ma, Xiaofei Xie, Yuejun Guo, Yves Le Traon","submitted_at":"2022-04-08T11:19:16Z","abstract_excerpt":"Deep Neural Networks (DNNs) have gained considerable attention in the past decades due to their astounding performance in different applications, such as natural language modeling, self-driving assistance, and source code understanding. With rapid exploration, more and more complex DNN architectures have been proposed along with huge pre-trained model parameters. The common way to use such DNN models in user-friendly devices (e.g., mobile phones) is to perform model compression before deployment. However, recent research has demonstrated that model compression, e.g., model quantization, yields"},"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":"2204.04220","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-04-08T11:19:16Z","cross_cats_sorted":["cs.AI","cs.SE"],"title_canon_sha256":"40487e9a05c37da18b1d277dddbbdfe106fc475a44ecabab4575b1ca79843469","abstract_canon_sha256":"ca1316e8e307b8625b9321446d7458403f96f7809f364adb71f905a5095eefd1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:12:54.575096Z","signature_b64":"F7t47YkgKqaZJA1GElqBLZVUm4jrV7vV6tCYJ7J32J4HE7WXHL8vrfcukVex06jEtSCw+Ny41LdhYrXqnbJoAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"96784abac1ba02dba719ff866dfe644319d825135930296c99cd21cd41604653","last_reissued_at":"2026-07-05T04:12:54.574667Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:12:54.574667Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Characterizing and Understanding the Behavior of Quantized Models for Reliable Deployment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SE"],"primary_cat":"cs.LG","authors_text":"Maxime Cordy, Mike Papadakis, Qiang Hu, Wei Ma, Xiaofei Xie, Yuejun Guo, Yves Le Traon","submitted_at":"2022-04-08T11:19:16Z","abstract_excerpt":"Deep Neural Networks (DNNs) have gained considerable attention in the past decades due to their astounding performance in different applications, such as natural language modeling, self-driving assistance, and source code understanding. With rapid exploration, more and more complex DNN architectures have been proposed along with huge pre-trained model parameters. The common way to use such DNN models in user-friendly devices (e.g., mobile phones) is to perform model compression before deployment. However, recent research has demonstrated that model compression, e.g., model quantization, yields"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.04220","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/2204.04220/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":"2204.04220","created_at":"2026-07-05T04:12:54.574723+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.04220v1","created_at":"2026-07-05T04:12:54.574723+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.04220","created_at":"2026-07-05T04:12:54.574723+00:00"},{"alias_kind":"pith_short_12","alias_value":"SZ4EVOWBXIBN","created_at":"2026-07-05T04:12:54.574723+00:00"},{"alias_kind":"pith_short_16","alias_value":"SZ4EVOWBXIBNXJYZ","created_at":"2026-07-05T04:12:54.574723+00:00"},{"alias_kind":"pith_short_8","alias_value":"SZ4EVOWB","created_at":"2026-07-05T04:12:54.574723+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.00859","citing_title":"Quantization Meets OOD: Generalizable Quantization-aware Training from a Flatness Perspective","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM","json":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM.json","graph_json":"https://pith.science/api/pith-number/SZ4EVOWBXIBNXJYZ76DG37TEIM/graph.json","events_json":"https://pith.science/api/pith-number/SZ4EVOWBXIBNXJYZ76DG37TEIM/events.json","paper":"https://pith.science/paper/SZ4EVOWB"},"agent_actions":{"view_html":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM","download_json":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM.json","view_paper":"https://pith.science/paper/SZ4EVOWB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.04220&json=true","fetch_graph":"https://pith.science/api/pith-number/SZ4EVOWBXIBNXJYZ76DG37TEIM/graph.json","fetch_events":"https://pith.science/api/pith-number/SZ4EVOWBXIBNXJYZ76DG37TEIM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM/action/storage_attestation","attest_author":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM/action/author_attestation","sign_citation":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM/action/citation_signature","submit_replication":"https://pith.science/pith/SZ4EVOWBXIBNXJYZ76DG37TEIM/action/replication_record"}},"created_at":"2026-07-05T04:12:54.574723+00:00","updated_at":"2026-07-05T04:12:54.574723+00:00"}