{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:MZYFDP74SWRXXFHXVPSJUY6WDR","short_pith_number":"pith:MZYFDP74","schema_version":"1.0","canonical_sha256":"667051bffc95a37b94f7abe49a63d61c451f7b7113aca8c9c6c266baaa0e7fc0","source":{"kind":"arxiv","id":"2310.19449","version":1},"attestation_state":"computed","paper":{"title":"Large-Scale Application of Fault Injection into PyTorch Models -- an Extension to PyTorchFI for Validation Efficiency","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Florian Geissler, Michael Paulitsch, Qutub Syed Sha, Ralf Graafe","submitted_at":"2023-10-30T11:18:35Z","abstract_excerpt":"Transient or permanent faults in hardware can render the output of Neural Networks (NN) incorrect without user-specific traces of the error, i.e. silent data errors (SDE). On the other hand, modern NNs also possess an inherent redundancy that can tolerate specific faults. To establish a safety case, it is necessary to distinguish and quantify both types of corruptions. To study the effects of hardware (HW) faults on software (SW) in general and NN models in particular, several fault injection (FI) methods have been established in recent years. Current FI methods focus on the methodology of inj"},"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":"2310.19449","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-10-30T11:18:35Z","cross_cats_sorted":[],"title_canon_sha256":"79e6cb05ca6721a28cd8873f1e22ef0fb5e9d2821fbba7e944c76f9ddda8d69d","abstract_canon_sha256":"37981b1a576f1d847ece80fcbb77ed3b95789c3b4ede34e787550ba29361d489"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:06:51.179448Z","signature_b64":"6dngjv3VJPm/GWN/yD11zvnWWCUCxgMBnobxPD+7WkByVas8TTVUny0/KB/gveb2KsHYaS3kotrgnsduP3DbCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"667051bffc95a37b94f7abe49a63d61c451f7b7113aca8c9c6c266baaa0e7fc0","last_reissued_at":"2026-07-05T07:06:51.179035Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:06:51.179035Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large-Scale Application of Fault Injection into PyTorch Models -- an Extension to PyTorchFI for Validation Efficiency","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Florian Geissler, Michael Paulitsch, Qutub Syed Sha, Ralf Graafe","submitted_at":"2023-10-30T11:18:35Z","abstract_excerpt":"Transient or permanent faults in hardware can render the output of Neural Networks (NN) incorrect without user-specific traces of the error, i.e. silent data errors (SDE). On the other hand, modern NNs also possess an inherent redundancy that can tolerate specific faults. To establish a safety case, it is necessary to distinguish and quantify both types of corruptions. To study the effects of hardware (HW) faults on software (SW) in general and NN models in particular, several fault injection (FI) methods have been established in recent years. Current FI methods focus on the methodology of inj"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.19449","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/2310.19449/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":"2310.19449","created_at":"2026-07-05T07:06:51.179095+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.19449v1","created_at":"2026-07-05T07:06:51.179095+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.19449","created_at":"2026-07-05T07:06:51.179095+00:00"},{"alias_kind":"pith_short_12","alias_value":"MZYFDP74SWRX","created_at":"2026-07-05T07:06:51.179095+00:00"},{"alias_kind":"pith_short_16","alias_value":"MZYFDP74SWRXXFHX","created_at":"2026-07-05T07:06:51.179095+00:00"},{"alias_kind":"pith_short_8","alias_value":"MZYFDP74","created_at":"2026-07-05T07:06:51.179095+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/MZYFDP74SWRXXFHXVPSJUY6WDR","json":"https://pith.science/pith/MZYFDP74SWRXXFHXVPSJUY6WDR.json","graph_json":"https://pith.science/api/pith-number/MZYFDP74SWRXXFHXVPSJUY6WDR/graph.json","events_json":"https://pith.science/api/pith-number/MZYFDP74SWRXXFHXVPSJUY6WDR/events.json","paper":"https://pith.science/paper/MZYFDP74"},"agent_actions":{"view_html":"https://pith.science/pith/MZYFDP74SWRXXFHXVPSJUY6WDR","download_json":"https://pith.science/pith/MZYFDP74SWRXXFHXVPSJUY6WDR.json","view_paper":"https://pith.science/paper/MZYFDP74","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.19449&json=true","fetch_graph":"https://pith.science/api/pith-number/MZYFDP74SWRXXFHXVPSJUY6WDR/graph.json","fetch_events":"https://pith.science/api/pith-number/MZYFDP74SWRXXFHXVPSJUY6WDR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MZYFDP74SWRXXFHXVPSJUY6WDR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MZYFDP74SWRXXFHXVPSJUY6WDR/action/storage_attestation","attest_author":"https://pith.science/pith/MZYFDP74SWRXXFHXVPSJUY6WDR/action/author_attestation","sign_citation":"https://pith.science/pith/MZYFDP74SWRXXFHXVPSJUY6WDR/action/citation_signature","submit_replication":"https://pith.science/pith/MZYFDP74SWRXXFHXVPSJUY6WDR/action/replication_record"}},"created_at":"2026-07-05T07:06:51.179095+00:00","updated_at":"2026-07-05T07:06:51.179095+00:00"}