{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:QRYWRLKHNMBBACAZ6XIE4IT45A","short_pith_number":"pith:QRYWRLKH","schema_version":"1.0","canonical_sha256":"847168ad476b02100819f5d04e227ce83ebd248a058f9bcfe12dcddbc5ce688f","source":{"kind":"arxiv","id":"2111.06934","version":1},"attestation_state":"computed","paper":{"title":"Contrastive Feature Loss for Image Prediction","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Alex Andonian, Bryan Russell, Jun-Yan Zhu, Phillip Isola, Richard Zhang, Taesung Park","submitted_at":"2021-11-12T20:39:52Z","abstract_excerpt":"Training supervised image synthesis models requires a critic to compare two images: the ground truth to the result. Yet, this basic functionality remains an open problem. A popular line of approaches uses the L1 (mean absolute error) loss, either in the pixel or the feature space of pretrained deep networks. However, we observe that these losses tend to produce overly blurry and grey images, and other techniques such as GANs need to be employed to fight these artifacts. In this work, we introduce an information theory based approach to measuring similarity between two images. We argue that a g"},"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":"2111.06934","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2021-11-12T20:39:52Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"dcf7ea529e3ca5d77cb7771de24ad3be15be070ba60e6e95faffcdeba4278dc5","abstract_canon_sha256":"e152d3f8369fb9a939543a865ba2c73dd8feb29cd5110b90c352bc9ca9200519"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:31:33.211108Z","signature_b64":"y9w9QDG3Z8cuEPm+ZeY3QeENzIAsHRLpoN1KM5Z/DCqVdScd2C7JCqVlgEVDuzoCdLITX6+cU3ZkpiQR9ODQCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"847168ad476b02100819f5d04e227ce83ebd248a058f9bcfe12dcddbc5ce688f","last_reissued_at":"2026-07-05T03:31:33.210751Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:31:33.210751Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contrastive Feature Loss for Image Prediction","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Alex Andonian, Bryan Russell, Jun-Yan Zhu, Phillip Isola, Richard Zhang, Taesung Park","submitted_at":"2021-11-12T20:39:52Z","abstract_excerpt":"Training supervised image synthesis models requires a critic to compare two images: the ground truth to the result. Yet, this basic functionality remains an open problem. A popular line of approaches uses the L1 (mean absolute error) loss, either in the pixel or the feature space of pretrained deep networks. However, we observe that these losses tend to produce overly blurry and grey images, and other techniques such as GANs need to be employed to fight these artifacts. In this work, we introduce an information theory based approach to measuring similarity between two images. We argue that a g"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2111.06934","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/2111.06934/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":"2111.06934","created_at":"2026-07-05T03:31:33.210818+00:00"},{"alias_kind":"arxiv_version","alias_value":"2111.06934v1","created_at":"2026-07-05T03:31:33.210818+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2111.06934","created_at":"2026-07-05T03:31:33.210818+00:00"},{"alias_kind":"pith_short_12","alias_value":"QRYWRLKHNMBB","created_at":"2026-07-05T03:31:33.210818+00:00"},{"alias_kind":"pith_short_16","alias_value":"QRYWRLKHNMBBACAZ","created_at":"2026-07-05T03:31:33.210818+00:00"},{"alias_kind":"pith_short_8","alias_value":"QRYWRLKH","created_at":"2026-07-05T03:31:33.210818+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/QRYWRLKHNMBBACAZ6XIE4IT45A","json":"https://pith.science/pith/QRYWRLKHNMBBACAZ6XIE4IT45A.json","graph_json":"https://pith.science/api/pith-number/QRYWRLKHNMBBACAZ6XIE4IT45A/graph.json","events_json":"https://pith.science/api/pith-number/QRYWRLKHNMBBACAZ6XIE4IT45A/events.json","paper":"https://pith.science/paper/QRYWRLKH"},"agent_actions":{"view_html":"https://pith.science/pith/QRYWRLKHNMBBACAZ6XIE4IT45A","download_json":"https://pith.science/pith/QRYWRLKHNMBBACAZ6XIE4IT45A.json","view_paper":"https://pith.science/paper/QRYWRLKH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2111.06934&json=true","fetch_graph":"https://pith.science/api/pith-number/QRYWRLKHNMBBACAZ6XIE4IT45A/graph.json","fetch_events":"https://pith.science/api/pith-number/QRYWRLKHNMBBACAZ6XIE4IT45A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QRYWRLKHNMBBACAZ6XIE4IT45A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QRYWRLKHNMBBACAZ6XIE4IT45A/action/storage_attestation","attest_author":"https://pith.science/pith/QRYWRLKHNMBBACAZ6XIE4IT45A/action/author_attestation","sign_citation":"https://pith.science/pith/QRYWRLKHNMBBACAZ6XIE4IT45A/action/citation_signature","submit_replication":"https://pith.science/pith/QRYWRLKHNMBBACAZ6XIE4IT45A/action/replication_record"}},"created_at":"2026-07-05T03:31:33.210818+00:00","updated_at":"2026-07-05T03:31:33.210818+00:00"}