{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:MK2EBWDFOR75K7DGJGWA7RJBB6","short_pith_number":"pith:MK2EBWDF","schema_version":"1.0","canonical_sha256":"62b440d865747fd57c6649ac0fc5210f8e5d46442070a9dc5b9d174e5a5ca12c","source":{"kind":"arxiv","id":"2403.19776","version":2},"attestation_state":"computed","paper":{"title":"Contrastive Test-Time Composition of Multiple LoRA Models for Image Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Enis Simsar, Federico Tombari, Pinar Yanardag, Tuna Han Salih Meral","submitted_at":"2024-03-28T18:58:43Z","abstract_excerpt":"Low-Rank Adaptation (LoRA) has emerged as a powerful and popular technique for personalization, enabling efficient adaptation of pre-trained image generation models for specific tasks without comprehensive retraining. While employing individual pre-trained LoRA models excels at representing single concepts, such as those representing a specific dog or a cat, utilizing multiple LoRA models to capture a variety of concepts in a single image still poses a significant challenge. Existing methods often fall short, primarily because the attention mechanisms within different LoRA models overlap, lead"},"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":"2403.19776","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-03-28T18:58:43Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"bd3db22d422f9a60e85ceed12e6f4ca3cc2fce96a489ba015bca449e148fc4f0","abstract_canon_sha256":"2c3ddb9265032263195868e8b51e28f096406c60d18ba5bfd4c611d8fda67476"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:45:12.941213Z","signature_b64":"wWfP7R7MXzMRmyYoY0LlvkDtmmY2/EKDxCHRuXRVC8AhYYxNhd2lyginNSNIX8Ru9A9sm3njyTQHVnPHW++iDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"62b440d865747fd57c6649ac0fc5210f8e5d46442070a9dc5b9d174e5a5ca12c","last_reissued_at":"2026-07-05T11:45:12.940743Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:45:12.940743Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Contrastive Test-Time Composition of Multiple LoRA Models for Image Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Enis Simsar, Federico Tombari, Pinar Yanardag, Tuna Han Salih Meral","submitted_at":"2024-03-28T18:58:43Z","abstract_excerpt":"Low-Rank Adaptation (LoRA) has emerged as a powerful and popular technique for personalization, enabling efficient adaptation of pre-trained image generation models for specific tasks without comprehensive retraining. While employing individual pre-trained LoRA models excels at representing single concepts, such as those representing a specific dog or a cat, utilizing multiple LoRA models to capture a variety of concepts in a single image still poses a significant challenge. Existing methods often fall short, primarily because the attention mechanisms within different LoRA models overlap, lead"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.19776","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/2403.19776/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":"2403.19776","created_at":"2026-07-05T11:45:12.940803+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.19776v2","created_at":"2026-07-05T11:45:12.940803+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.19776","created_at":"2026-07-05T11:45:12.940803+00:00"},{"alias_kind":"pith_short_12","alias_value":"MK2EBWDFOR75","created_at":"2026-07-05T11:45:12.940803+00:00"},{"alias_kind":"pith_short_16","alias_value":"MK2EBWDFOR75K7DG","created_at":"2026-07-05T11:45:12.940803+00:00"},{"alias_kind":"pith_short_8","alias_value":"MK2EBWDF","created_at":"2026-07-05T11:45:12.940803+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.01792","citing_title":"FreeLoRA: Enabling Training-Free LoRA Fusion for Autoregressive Multi-Subject Personalization","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6","json":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6.json","graph_json":"https://pith.science/api/pith-number/MK2EBWDFOR75K7DGJGWA7RJBB6/graph.json","events_json":"https://pith.science/api/pith-number/MK2EBWDFOR75K7DGJGWA7RJBB6/events.json","paper":"https://pith.science/paper/MK2EBWDF"},"agent_actions":{"view_html":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6","download_json":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6.json","view_paper":"https://pith.science/paper/MK2EBWDF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.19776&json=true","fetch_graph":"https://pith.science/api/pith-number/MK2EBWDFOR75K7DGJGWA7RJBB6/graph.json","fetch_events":"https://pith.science/api/pith-number/MK2EBWDFOR75K7DGJGWA7RJBB6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6/action/storage_attestation","attest_author":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6/action/author_attestation","sign_citation":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6/action/citation_signature","submit_replication":"https://pith.science/pith/MK2EBWDFOR75K7DGJGWA7RJBB6/action/replication_record"}},"created_at":"2026-07-05T11:45:12.940803+00:00","updated_at":"2026-07-05T11:45:12.940803+00:00"}