{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:UGLRDB77IOHVVDYNQAYXLRLE5G","short_pith_number":"pith:UGLRDB77","canonical_record":{"source":{"id":"2508.06202","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-08T10:32:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1d2065b36cfb8a9dd20d2aa5bb7963ee68228d39536ff17c78c76c6572b8eef8","abstract_canon_sha256":"108946eef16c771caaade33032e0832e1e36a35fce39d631d1154c937d54d115"},"schema_version":"1.0"},"canonical_sha256":"a1971187ff438f5a8f0d803175c564e9a13537a69605a3fa9f736e20f6e4c70c","source":{"kind":"arxiv","id":"2508.06202","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06202","created_at":"2026-05-17T23:39:17Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06202v2","created_at":"2026-05-17T23:39:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06202","created_at":"2026-05-17T23:39:17Z"},{"alias_kind":"pith_short_12","alias_value":"UGLRDB77IOHV","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"UGLRDB77IOHVVDYN","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"UGLRDB77","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:UGLRDB77IOHVVDYNQAYXLRLE5G","target":"record","payload":{"canonical_record":{"source":{"id":"2508.06202","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-08T10:32:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1d2065b36cfb8a9dd20d2aa5bb7963ee68228d39536ff17c78c76c6572b8eef8","abstract_canon_sha256":"108946eef16c771caaade33032e0832e1e36a35fce39d631d1154c937d54d115"},"schema_version":"1.0"},"canonical_sha256":"a1971187ff438f5a8f0d803175c564e9a13537a69605a3fa9f736e20f6e4c70c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:17.757536Z","signature_b64":"2s5d1Sn2yRBzhMeqDBxlncyfh0vimUpxsBWdSC5AFX3eMALKfw9a4ni7h3jNpYbVu9Q5Lf9AEyFPuDz8huzHAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a1971187ff438f5a8f0d803175c564e9a13537a69605a3fa9f736e20f6e4c70c","last_reissued_at":"2026-05-17T23:39:17.756893Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:17.756893Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.06202","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hOAa8lo8g8xV2aNgeYsZnPr+P74u29SAfEp8cxomlIXoMfiSPhFla/QvrP8jYTtOv4OCDceggQ0lpfCLsXPTCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T14:47:30.303755Z"},"content_sha256":"5b5beef72206929889270fcb4b3c447a87c5308009662fe05927f871ba21989c","schema_version":"1.0","event_id":"sha256:5b5beef72206929889270fcb4b3c447a87c5308009662fe05927f871ba21989c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:UGLRDB77IOHVVDYNQAYXLRLE5G","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LoRA in LoRA: Towards Parameter-Efficient Architecture Expansion for Continual Visual Instruction Tuning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chang Che, Hui Ma, Pengwan Yang, Qi Wang, Zenglin Shi, Ziqi Wang","submitted_at":"2025-08-08T10:32:38Z","abstract_excerpt":"Continual Visual Instruction Tuning (CVIT) enables Multimodal Large Language Models (MLLMs) to incrementally learn new tasks over time. However, this process is challenged by catastrophic forgetting, where performance on previously learned tasks deteriorates as the model adapts to new ones. A common approach to mitigate forgetting is architecture expansion, which introduces task-specific modules to prevent interference. Yet, existing methods often expand entire layers for each task, leading to significant parameter overhead and poor scalability. To overcome these issues, we introduce LoRA in L"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06202","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":""},"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":"2026-05-17T23:39:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pnqMAStOwjlHR3BUGrCw//97jLBi4bGeX6EVT/IUE7nADDPYZff6sbHXnzS3oTNso4Bylgm3+Mbz1UTtswXABg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T14:47:30.304110Z"},"content_sha256":"a766ac7fafcf0c907a9e1815e63a018916e16b2463fb39cbad0c72417449cd2b","schema_version":"1.0","event_id":"sha256:a766ac7fafcf0c907a9e1815e63a018916e16b2463fb39cbad0c72417449cd2b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UGLRDB77IOHVVDYNQAYXLRLE5G/bundle.json","state_url":"https://pith.science/pith/UGLRDB77IOHVVDYNQAYXLRLE5G/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UGLRDB77IOHVVDYNQAYXLRLE5G/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":"2026-05-21T14:47:30Z","links":{"resolver":"https://pith.science/pith/UGLRDB77IOHVVDYNQAYXLRLE5G","bundle":"https://pith.science/pith/UGLRDB77IOHVVDYNQAYXLRLE5G/bundle.json","state":"https://pith.science/pith/UGLRDB77IOHVVDYNQAYXLRLE5G/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UGLRDB77IOHVVDYNQAYXLRLE5G/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UGLRDB77IOHVVDYNQAYXLRLE5G","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":"108946eef16c771caaade33032e0832e1e36a35fce39d631d1154c937d54d115","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-08T10:32:38Z","title_canon_sha256":"1d2065b36cfb8a9dd20d2aa5bb7963ee68228d39536ff17c78c76c6572b8eef8"},"schema_version":"1.0","source":{"id":"2508.06202","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.06202","created_at":"2026-05-17T23:39:17Z"},{"alias_kind":"arxiv_version","alias_value":"2508.06202v2","created_at":"2026-05-17T23:39:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06202","created_at":"2026-05-17T23:39:17Z"},{"alias_kind":"pith_short_12","alias_value":"UGLRDB77IOHV","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"UGLRDB77IOHVVDYN","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"UGLRDB77","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:a766ac7fafcf0c907a9e1815e63a018916e16b2463fb39cbad0c72417449cd2b","target":"graph","created_at":"2026-05-17T23:39:17Z","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"},"paper":{"abstract_excerpt":"Continual Visual Instruction Tuning (CVIT) enables Multimodal Large Language Models (MLLMs) to incrementally learn new tasks over time. However, this process is challenged by catastrophic forgetting, where performance on previously learned tasks deteriorates as the model adapts to new ones. A common approach to mitigate forgetting is architecture expansion, which introduces task-specific modules to prevent interference. Yet, existing methods often expand entire layers for each task, leading to significant parameter overhead and poor scalability. To overcome these issues, we introduce LoRA in L","authors_text":"Chang Che, Hui Ma, Pengwan Yang, Qi Wang, Zenglin Shi, Ziqi Wang","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-08T10:32:38Z","title":"LoRA in LoRA: Towards Parameter-Efficient Architecture Expansion for Continual Visual Instruction Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06202","kind":"arxiv","version":2},"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":"sha256:5b5beef72206929889270fcb4b3c447a87c5308009662fe05927f871ba21989c","target":"record","created_at":"2026-05-17T23:39:17Z","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":"108946eef16c771caaade33032e0832e1e36a35fce39d631d1154c937d54d115","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-08T10:32:38Z","title_canon_sha256":"1d2065b36cfb8a9dd20d2aa5bb7963ee68228d39536ff17c78c76c6572b8eef8"},"schema_version":"1.0","source":{"id":"2508.06202","kind":"arxiv","version":2}},"canonical_sha256":"a1971187ff438f5a8f0d803175c564e9a13537a69605a3fa9f736e20f6e4c70c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a1971187ff438f5a8f0d803175c564e9a13537a69605a3fa9f736e20f6e4c70c","first_computed_at":"2026-05-17T23:39:17.756893Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:17.756893Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2s5d1Sn2yRBzhMeqDBxlncyfh0vimUpxsBWdSC5AFX3eMALKfw9a4ni7h3jNpYbVu9Q5Lf9AEyFPuDz8huzHAQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:17.757536Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.06202","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5b5beef72206929889270fcb4b3c447a87c5308009662fe05927f871ba21989c","sha256:a766ac7fafcf0c907a9e1815e63a018916e16b2463fb39cbad0c72417449cd2b"],"state_sha256":"e25c8bd8c3b60a6f317eff2f96548d7c45457996f034dd9f03b2ba30cc84137b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"908ydp+GvtY5veH2TrmxH8kge12V+jJVoLH/CKFQgf5zikOg496D0dUYbnqF6qQJlCORSfj+0Ix244c/MEb8BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-21T14:47:30.306141Z","bundle_sha256":"1042b9f925dad4366e90ead8d36668f5e8957674f8b8dd2964c2d9d263b3df54"}}