{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PIJIOO54LVDQQBS4RLMPF75XEQ","short_pith_number":"pith:PIJIOO54","schema_version":"1.0","canonical_sha256":"7a12873bbc5d4708065c8ad8f2ffb72406c05056cfb9b990c085340babbef7c1","source":{"kind":"arxiv","id":"2408.06741","version":2},"attestation_state":"computed","paper":{"title":"Improving Synthetic Image Detection Towards Generalization: An Image Transformation Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fuli Feng, Jiayin Cai, Ouxiang Li, Xiaolong Jiang, Yanbin Hao, Yao Hu","submitted_at":"2024-08-13T09:01:12Z","abstract_excerpt":"With recent generative models facilitating photo-realistic image synthesis, the proliferation of synthetic images has also engendered certain negative impacts on social platforms, thereby raising an urgent imperative to develop effective detectors. Current synthetic image detection (SID) pipelines are primarily dedicated to crafting universal artifact features, accompanied by an oversight about SID training paradigm. In this paper, we re-examine the SID problem and identify two prevalent biases in current training paradigms, i.e., weakened artifact features and overfitted artifact features. Me"},"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":"2408.06741","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-08-13T09:01:12Z","cross_cats_sorted":[],"title_canon_sha256":"d4b0b5e84fc908f458b9f3b1cd3fa44c841d76134f47c8e15d267dcfa417327d","abstract_canon_sha256":"a4213889e3052dbc5a487633c80ea72bb2cd22480c181a4abf0a1e9aeb0c0418"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:56:53.762829Z","signature_b64":"WD5tvrBm5kfnux0ffMfAJcT2VeZrQ0xFawWQlLCxHTvpP5J3HKK54BNMTGDB7/M3e4IBlBiCYdC+vUdoCLe3CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a12873bbc5d4708065c8ad8f2ffb72406c05056cfb9b990c085340babbef7c1","last_reissued_at":"2026-07-05T09:56:53.762393Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:56:53.762393Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Synthetic Image Detection Towards Generalization: An Image Transformation Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fuli Feng, Jiayin Cai, Ouxiang Li, Xiaolong Jiang, Yanbin Hao, Yao Hu","submitted_at":"2024-08-13T09:01:12Z","abstract_excerpt":"With recent generative models facilitating photo-realistic image synthesis, the proliferation of synthetic images has also engendered certain negative impacts on social platforms, thereby raising an urgent imperative to develop effective detectors. Current synthetic image detection (SID) pipelines are primarily dedicated to crafting universal artifact features, accompanied by an oversight about SID training paradigm. In this paper, we re-examine the SID problem and identify two prevalent biases in current training paradigms, i.e., weakened artifact features and overfitted artifact features. Me"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.06741","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/2408.06741/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":"2408.06741","created_at":"2026-07-05T09:56:53.762449+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.06741v2","created_at":"2026-07-05T09:56:53.762449+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.06741","created_at":"2026-07-05T09:56:53.762449+00:00"},{"alias_kind":"pith_short_12","alias_value":"PIJIOO54LVDQ","created_at":"2026-07-05T09:56:53.762449+00:00"},{"alias_kind":"pith_short_16","alias_value":"PIJIOO54LVDQQBS4","created_at":"2026-07-05T09:56:53.762449+00:00"},{"alias_kind":"pith_short_8","alias_value":"PIJIOO54","created_at":"2026-07-05T09:56:53.762449+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08634","citing_title":"SSAFE: Simple and Strong AI-Generated Image Detection via Frozen Vision Encoders","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27348","citing_title":"When Eyes Betray AI: Social Gaze Consistency as a Semantic Cue for AI-Generated Image Detection","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2502.19716","citing_title":"Fully AI-Generated Image Detection: Definition, Recent Advances and Challenges","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2504.18361","citing_title":"COCO-Inpaint: A Benchmark for Detecting and Localizing Inpainting-Based Image Manipulations","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ","json":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ.json","graph_json":"https://pith.science/api/pith-number/PIJIOO54LVDQQBS4RLMPF75XEQ/graph.json","events_json":"https://pith.science/api/pith-number/PIJIOO54LVDQQBS4RLMPF75XEQ/events.json","paper":"https://pith.science/paper/PIJIOO54"},"agent_actions":{"view_html":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ","download_json":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ.json","view_paper":"https://pith.science/paper/PIJIOO54","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.06741&json=true","fetch_graph":"https://pith.science/api/pith-number/PIJIOO54LVDQQBS4RLMPF75XEQ/graph.json","fetch_events":"https://pith.science/api/pith-number/PIJIOO54LVDQQBS4RLMPF75XEQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ/action/storage_attestation","attest_author":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ/action/author_attestation","sign_citation":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ/action/citation_signature","submit_replication":"https://pith.science/pith/PIJIOO54LVDQQBS4RLMPF75XEQ/action/replication_record"}},"created_at":"2026-07-05T09:56:53.762449+00:00","updated_at":"2026-07-05T09:56:53.762449+00:00"}