{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:Z4RCY3EP4BOT4BYBWDHWVWU5RM","short_pith_number":"pith:Z4RCY3EP","canonical_record":{"source":{"id":"2605.22743","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T17:13:49Z","cross_cats_sorted":[],"title_canon_sha256":"4275d35bd736f8a6c375f8658e89f2e6c53cd6eed9eec884ed0b708fb9204b78","abstract_canon_sha256":"38564f12bdb160055b248687fc71e47cf8f23c368136758c73f53bb2013f10d4"},"schema_version":"1.0"},"canonical_sha256":"cf222c6c8fe05d3e0701b0cf6ada9d8b0e3d120d5314c00426ab4b6c2c736d2d","source":{"kind":"arxiv","id":"2605.22743","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.22743","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"arxiv_version","alias_value":"2605.22743v1","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22743","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"pith_short_12","alias_value":"Z4RCY3EP4BOT","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"pith_short_16","alias_value":"Z4RCY3EP4BOT4BYB","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"pith_short_8","alias_value":"Z4RCY3EP","created_at":"2026-05-22T02:04:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:Z4RCY3EP4BOT4BYBWDHWVWU5RM","target":"record","payload":{"canonical_record":{"source":{"id":"2605.22743","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T17:13:49Z","cross_cats_sorted":[],"title_canon_sha256":"4275d35bd736f8a6c375f8658e89f2e6c53cd6eed9eec884ed0b708fb9204b78","abstract_canon_sha256":"38564f12bdb160055b248687fc71e47cf8f23c368136758c73f53bb2013f10d4"},"schema_version":"1.0"},"canonical_sha256":"cf222c6c8fe05d3e0701b0cf6ada9d8b0e3d120d5314c00426ab4b6c2c736d2d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T02:04:52.890453Z","signature_b64":"1NhhAWus7Suxzn7WnVRj4u2ChvzvLtQQYGr9gnZDp1vXX6O8sHII97punAQJKIKkjgaNZfET+XOwtMo1bz5wBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf222c6c8fe05d3e0701b0cf6ada9d8b0e3d120d5314c00426ab4b6c2c736d2d","last_reissued_at":"2026-05-22T02:04:52.889691Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T02:04:52.889691Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.22743","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T02:04:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e0mb4vF98zPI8rZV1886ppXH9RO/adHSBdU0VGSibWda3s/ufzb4ZyA2B90puvL9ZgFAfOAZ43UGrYCHkJHaCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T15:12:35.397169Z"},"content_sha256":"981b07c614bc2e6667584ad216314d42bcd2a75039468396c660df2bd2f31fd8","schema_version":"1.0","event_id":"sha256:981b07c614bc2e6667584ad216314d42bcd2a75039468396c660df2bd2f31fd8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:Z4RCY3EP4BOT4BYBWDHWVWU5RM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SeqLoRA: Bilevel Orthogonal Adaptation for Continual Multi-Concept Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amir Joudaki, Andr\\'e M. H. Teixeira, Enis Simsar, Javad Parsa, Thomas Hofmann","submitted_at":"2026-05-21T17:13:49Z","abstract_excerpt":"Parameter-efficient fine-tuning enables fast personalization of text-to-image diffusion models, but composing multiple custom concepts remains challenging due to representation interference. Existing modular methods either rely on expensive post-hoc fusion or freeze adaptation subspaces, which limit expressiveness and concept fidelity. To address this trade-off, we propose Sequential regularized LoRA (SeqLoRA), a constrained continual learning framework that jointly optimizes both LoRA factors via bilevel optimization. Theoretically, we establish strong convergence guarantees for our algorithm"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22743","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/2605.22743/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T02:04:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KCt2AoNoW6PgK4LtGiaueDYwOkq7zaQ9lGyUXxnAwEJdBzObhrqhW1l/Wvs0+jiIkAYVYtoXFmSR9A+VudXQAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-22T15:12:35.397653Z"},"content_sha256":"570148d393d6acda0b79d9b9fa01a4e1feea1c170ee46eb81551734d69c04397","schema_version":"1.0","event_id":"sha256:570148d393d6acda0b79d9b9fa01a4e1feea1c170ee46eb81551734d69c04397"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Z4RCY3EP4BOT4BYBWDHWVWU5RM/bundle.json","state_url":"https://pith.science/pith/Z4RCY3EP4BOT4BYBWDHWVWU5RM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Z4RCY3EP4BOT4BYBWDHWVWU5RM/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-22T15:12:35Z","links":{"resolver":"https://pith.science/pith/Z4RCY3EP4BOT4BYBWDHWVWU5RM","bundle":"https://pith.science/pith/Z4RCY3EP4BOT4BYBWDHWVWU5RM/bundle.json","state":"https://pith.science/pith/Z4RCY3EP4BOT4BYBWDHWVWU5RM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Z4RCY3EP4BOT4BYBWDHWVWU5RM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:Z4RCY3EP4BOT4BYBWDHWVWU5RM","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"38564f12bdb160055b248687fc71e47cf8f23c368136758c73f53bb2013f10d4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T17:13:49Z","title_canon_sha256":"4275d35bd736f8a6c375f8658e89f2e6c53cd6eed9eec884ed0b708fb9204b78"},"schema_version":"1.0","source":{"id":"2605.22743","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.22743","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"arxiv_version","alias_value":"2605.22743v1","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22743","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"pith_short_12","alias_value":"Z4RCY3EP4BOT","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"pith_short_16","alias_value":"Z4RCY3EP4BOT4BYB","created_at":"2026-05-22T02:04:52Z"},{"alias_kind":"pith_short_8","alias_value":"Z4RCY3EP","created_at":"2026-05-22T02:04:52Z"}],"graph_snapshots":[{"event_id":"sha256:570148d393d6acda0b79d9b9fa01a4e1feea1c170ee46eb81551734d69c04397","target":"graph","created_at":"2026-05-22T02:04:52Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.22743/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Parameter-efficient fine-tuning enables fast personalization of text-to-image diffusion models, but composing multiple custom concepts remains challenging due to representation interference. Existing modular methods either rely on expensive post-hoc fusion or freeze adaptation subspaces, which limit expressiveness and concept fidelity. To address this trade-off, we propose Sequential regularized LoRA (SeqLoRA), a constrained continual learning framework that jointly optimizes both LoRA factors via bilevel optimization. Theoretically, we establish strong convergence guarantees for our algorithm","authors_text":"Amir Joudaki, Andr\\'e M. H. Teixeira, Enis Simsar, Javad Parsa, Thomas Hofmann","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T17:13:49Z","title":"SeqLoRA: Bilevel Orthogonal Adaptation for Continual Multi-Concept Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22743","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:981b07c614bc2e6667584ad216314d42bcd2a75039468396c660df2bd2f31fd8","target":"record","created_at":"2026-05-22T02:04:52Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"38564f12bdb160055b248687fc71e47cf8f23c368136758c73f53bb2013f10d4","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-21T17:13:49Z","title_canon_sha256":"4275d35bd736f8a6c375f8658e89f2e6c53cd6eed9eec884ed0b708fb9204b78"},"schema_version":"1.0","source":{"id":"2605.22743","kind":"arxiv","version":1}},"canonical_sha256":"cf222c6c8fe05d3e0701b0cf6ada9d8b0e3d120d5314c00426ab4b6c2c736d2d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cf222c6c8fe05d3e0701b0cf6ada9d8b0e3d120d5314c00426ab4b6c2c736d2d","first_computed_at":"2026-05-22T02:04:52.889691Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T02:04:52.889691Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1NhhAWus7Suxzn7WnVRj4u2ChvzvLtQQYGr9gnZDp1vXX6O8sHII97punAQJKIKkjgaNZfET+XOwtMo1bz5wBA==","signature_status":"signed_v1","signed_at":"2026-05-22T02:04:52.890453Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.22743","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:981b07c614bc2e6667584ad216314d42bcd2a75039468396c660df2bd2f31fd8","sha256:570148d393d6acda0b79d9b9fa01a4e1feea1c170ee46eb81551734d69c04397"],"state_sha256":"258c15ebf277b91f59ab7e6a60654b1402542e283f32ff32a7f82bffbb88a31f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rVtkcv4Ev3nPxv6RrqWRbRmgtxyrUpHCXlgNY+mSNI+y922y5pSvwQDdN7bf7TMuDBVKmz41yM1tIc0wDUkGAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-22T15:12:35.401091Z","bundle_sha256":"aeb78b39904f1e019c9702046109bc0ace6846aaab6e95bdbc2d02b07ca70834"}}