{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:CGAWRI5OLD5FZEV5HDP2HW3B7C","short_pith_number":"pith:CGAWRI5O","schema_version":"1.0","canonical_sha256":"118168a3ae58fa5c92bd38dfa3db61f8a9b6b24bbcf9d322ab7c190cbc489c8a","source":{"kind":"arxiv","id":"2411.19261","version":2},"attestation_state":"computed","paper":{"title":"Improving Multi-Subject Consistency in Open-Domain Image Generation with Isolation and Reposition Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongyang Chao, Huan Yang, Huiguo He, Jian Yin, Qiuyue Wang, Yuan Zhou, Yuxuan Cai","submitted_at":"2024-11-28T16:50:30Z","abstract_excerpt":"Training-free diffusion models have achieved remarkable progress in generating multi-subject consistent images within open-domain scenarios. The key idea of these methods is to incorporate reference subject information within the attention layer. However, existing methods still obtain suboptimal performance when handling numerous subjects. This paper reveals two primary issues contributing to this deficiency. Firstly, the undesired internal attraction between different subjects within the target image can lead to the convergence of multiple subjects into a single entity. Secondly, tokens tend "},"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":"2411.19261","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-11-28T16:50:30Z","cross_cats_sorted":[],"title_canon_sha256":"a0394870cce7d09e0355afdfab560ee791a809b8df16230f94902aadc24e0791","abstract_canon_sha256":"53cda789492efbc664da9b4bb398208a681259075a37a8ebf9f7dae01768bc2b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:27:20.546572Z","signature_b64":"xhrI99wcsS9COKM6qatIUQvypeUWI4gycjfXAowmksY+DOfMMaH5PAAz18M2k+WuTrjQyAAfrdAuYBJlBWc4BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"118168a3ae58fa5c92bd38dfa3db61f8a9b6b24bbcf9d322ab7c190cbc489c8a","last_reissued_at":"2026-07-05T10:27:20.546063Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:27:20.546063Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Multi-Subject Consistency in Open-Domain Image Generation with Isolation and Reposition Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongyang Chao, Huan Yang, Huiguo He, Jian Yin, Qiuyue Wang, Yuan Zhou, Yuxuan Cai","submitted_at":"2024-11-28T16:50:30Z","abstract_excerpt":"Training-free diffusion models have achieved remarkable progress in generating multi-subject consistent images within open-domain scenarios. The key idea of these methods is to incorporate reference subject information within the attention layer. However, existing methods still obtain suboptimal performance when handling numerous subjects. This paper reveals two primary issues contributing to this deficiency. Firstly, the undesired internal attraction between different subjects within the target image can lead to the convergence of multiple subjects into a single entity. Secondly, tokens tend "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.19261","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/2411.19261/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":"2411.19261","created_at":"2026-07-05T10:27:20.546122+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.19261v2","created_at":"2026-07-05T10:27:20.546122+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.19261","created_at":"2026-07-05T10:27:20.546122+00:00"},{"alias_kind":"pith_short_12","alias_value":"CGAWRI5OLD5F","created_at":"2026-07-05T10:27:20.546122+00:00"},{"alias_kind":"pith_short_16","alias_value":"CGAWRI5OLD5FZEV5","created_at":"2026-07-05T10:27:20.546122+00:00"},{"alias_kind":"pith_short_8","alias_value":"CGAWRI5O","created_at":"2026-07-05T10:27:20.546122+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.04123","citing_title":"TaleDiffusion: Multi-Character Story Generation with Dialogue Rendering","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2512.08477","citing_title":"ContextDrag: Precise Drag-Based Image Editing via Context-Preserving Token Injection and Position-Aligned Attention","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C","json":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C.json","graph_json":"https://pith.science/api/pith-number/CGAWRI5OLD5FZEV5HDP2HW3B7C/graph.json","events_json":"https://pith.science/api/pith-number/CGAWRI5OLD5FZEV5HDP2HW3B7C/events.json","paper":"https://pith.science/paper/CGAWRI5O"},"agent_actions":{"view_html":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C","download_json":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C.json","view_paper":"https://pith.science/paper/CGAWRI5O","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.19261&json=true","fetch_graph":"https://pith.science/api/pith-number/CGAWRI5OLD5FZEV5HDP2HW3B7C/graph.json","fetch_events":"https://pith.science/api/pith-number/CGAWRI5OLD5FZEV5HDP2HW3B7C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C/action/storage_attestation","attest_author":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C/action/author_attestation","sign_citation":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C/action/citation_signature","submit_replication":"https://pith.science/pith/CGAWRI5OLD5FZEV5HDP2HW3B7C/action/replication_record"}},"created_at":"2026-07-05T10:27:20.546122+00:00","updated_at":"2026-07-05T10:27:20.546122+00:00"}