{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MKHTGQH4G2NREW67L37HO3O4X3","short_pith_number":"pith:MKHTGQH4","schema_version":"1.0","canonical_sha256":"628f3340fc369b125bdf5efe776ddcbed48e652de95f0b254f4764f60bcac115","source":{"kind":"arxiv","id":"2405.06914","version":1},"attestation_state":"computed","paper":{"title":"Non-confusing Generation of Customized Concepts in Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Liang, Fei Wu, Hanwang Zhang, Jiaxin Shi, Jingyuan Chen, Junzhong Miao, Shuicheng Yan, Tao Jin, Wang Lin, Yichen Zhu, Zhou Zhao","submitted_at":"2024-05-11T05:01:53Z","abstract_excerpt":"We tackle the common challenge of inter-concept visual confusion in compositional concept generation using text-guided diffusion models (TGDMs). It becomes even more pronounced in the generation of customized concepts, due to the scarcity of user-provided concept visual examples. By revisiting the two major stages leading to the success of TGDMs -- 1) contrastive image-language pre-training (CLIP) for text encoder that encodes visual semantics, and 2) training TGDM that decodes the textual embeddings into pixels -- we point that existing customized generation methods only focus on fine-tuning "},"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":"2405.06914","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-05-11T05:01:53Z","cross_cats_sorted":[],"title_canon_sha256":"9be727990ad70b207153377f48b587309250cd0f91f7ebef55b185c1d9242cda","abstract_canon_sha256":"72ab4ab5a69133743fdcf3a96dbe1597dc8afed892ccb562211a2f65a61ab42d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:18:13.757968Z","signature_b64":"tnpOjw5uDy98f1RFRMDGbZdOnOVkpweuAGg6GDzfCyxnWRwPL+ieGWzcm+L32IkNxaBmVbPE4m4ZD/LowJBSDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"628f3340fc369b125bdf5efe776ddcbed48e652de95f0b254f4764f60bcac115","last_reissued_at":"2026-07-05T08:18:13.757480Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:18:13.757480Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Non-confusing Generation of Customized Concepts in Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Liang, Fei Wu, Hanwang Zhang, Jiaxin Shi, Jingyuan Chen, Junzhong Miao, Shuicheng Yan, Tao Jin, Wang Lin, Yichen Zhu, Zhou Zhao","submitted_at":"2024-05-11T05:01:53Z","abstract_excerpt":"We tackle the common challenge of inter-concept visual confusion in compositional concept generation using text-guided diffusion models (TGDMs). It becomes even more pronounced in the generation of customized concepts, due to the scarcity of user-provided concept visual examples. By revisiting the two major stages leading to the success of TGDMs -- 1) contrastive image-language pre-training (CLIP) for text encoder that encodes visual semantics, and 2) training TGDM that decodes the textual embeddings into pixels -- we point that existing customized generation methods only focus on fine-tuning "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.06914","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/2405.06914/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":"2405.06914","created_at":"2026-07-05T08:18:13.757538+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.06914v1","created_at":"2026-07-05T08:18:13.757538+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.06914","created_at":"2026-07-05T08:18:13.757538+00:00"},{"alias_kind":"pith_short_12","alias_value":"MKHTGQH4G2NR","created_at":"2026-07-05T08:18:13.757538+00:00"},{"alias_kind":"pith_short_16","alias_value":"MKHTGQH4G2NREW67","created_at":"2026-07-05T08:18:13.757538+00:00"},{"alias_kind":"pith_short_8","alias_value":"MKHTGQH4","created_at":"2026-07-05T08:18:13.757538+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.07878","citing_title":"TAP: Parameter-efficient Task-Aware Prompting for Adverse Weather Removal","ref_index":39,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3","json":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3.json","graph_json":"https://pith.science/api/pith-number/MKHTGQH4G2NREW67L37HO3O4X3/graph.json","events_json":"https://pith.science/api/pith-number/MKHTGQH4G2NREW67L37HO3O4X3/events.json","paper":"https://pith.science/paper/MKHTGQH4"},"agent_actions":{"view_html":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3","download_json":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3.json","view_paper":"https://pith.science/paper/MKHTGQH4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.06914&json=true","fetch_graph":"https://pith.science/api/pith-number/MKHTGQH4G2NREW67L37HO3O4X3/graph.json","fetch_events":"https://pith.science/api/pith-number/MKHTGQH4G2NREW67L37HO3O4X3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3/action/storage_attestation","attest_author":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3/action/author_attestation","sign_citation":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3/action/citation_signature","submit_replication":"https://pith.science/pith/MKHTGQH4G2NREW67L37HO3O4X3/action/replication_record"}},"created_at":"2026-07-05T08:18:13.757538+00:00","updated_at":"2026-07-05T08:18:13.757538+00:00"}