{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:2MYR2HSV7CYX4VNAN26R6X3IYE","short_pith_number":"pith:2MYR2HSV","canonical_record":{"source":{"id":"2605.12752","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T21:06:03Z","cross_cats_sorted":[],"title_canon_sha256":"3e47e494325e05d37d6dd634f0e63dabbb1709d60463e7d92c67df390f757553","abstract_canon_sha256":"ece6f9960e995c19791123bd6a4d83755bd7db56e1331e3ac7c533225a36ecdb"},"schema_version":"1.0"},"canonical_sha256":"d3311d1e55f8b17e55a06ebd1f5f68c1025160fb359a75d8e4b80029c8ccd10d","source":{"kind":"arxiv","id":"2605.12752","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12752","created_at":"2026-05-18T03:09:48Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12752v1","created_at":"2026-05-18T03:09:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12752","created_at":"2026-05-18T03:09:48Z"},{"alias_kind":"pith_short_12","alias_value":"2MYR2HSV7CYX","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"2MYR2HSV7CYX4VNA","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"2MYR2HSV","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:2MYR2HSV7CYX4VNAN26R6X3IYE","target":"record","payload":{"canonical_record":{"source":{"id":"2605.12752","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T21:06:03Z","cross_cats_sorted":[],"title_canon_sha256":"3e47e494325e05d37d6dd634f0e63dabbb1709d60463e7d92c67df390f757553","abstract_canon_sha256":"ece6f9960e995c19791123bd6a4d83755bd7db56e1331e3ac7c533225a36ecdb"},"schema_version":"1.0"},"canonical_sha256":"d3311d1e55f8b17e55a06ebd1f5f68c1025160fb359a75d8e4b80029c8ccd10d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:48.735989Z","signature_b64":"zQjeWlrb4LWuZ7/06TgyvrPNOkPm8FlLpmL1eAdH+JCe1fpg++vQP2KYfVCaWZrxyRNC0JEq7REhEYkiUZgfBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d3311d1e55f8b17e55a06ebd1f5f68c1025160fb359a75d8e4b80029c8ccd10d","last_reissued_at":"2026-05-18T03:09:48.735110Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:48.735110Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.12752","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":"2026-05-18T03:09:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1wwZbVdY05nL422xxCQFzN9XZBWxevu6CPnnguhgNUewD/0tzofbzn3asQko66YDKjPl3BZ4Xy4VcNuS1M6tBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T15:51:18.327098Z"},"content_sha256":"1f4d5393c009b3489df10822c5d77fcc39ff8cf56b2473312647557e67700235","schema_version":"1.0","event_id":"sha256:1f4d5393c009b3489df10822c5d77fcc39ff8cf56b2473312647557e67700235"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:2MYR2HSV7CYX4VNAN26R6X3IYE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Low-Rank Adapters Initialization via Gradient Surgery for Continual Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"SLICE initializes LoRA adapters by projecting current and replay gradients then applying truncated SVD to reduce catastrophic forgetting in continual learning.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Arthur S. Bianchessi, Christian Mattjie, Joana Pasquali, Jo\\~ao Vitor Boer Abitante, Lucas S. Kupssinsk\\\"u, Ot\\'avio Parraga, Rafaela Cappelari Ravazio, Ramiro N. Barros, Rodrigo C. Barros, Vin\\'icius Conte Turani","submitted_at":"2026-05-12T21:06:03Z","abstract_excerpt":"LoRA is widely adopted for continual fine-tuning of Large Language Models due to its parameter efficiency, modularity across tasks, and compatibility with replay strategies. However, LoRA-based continual learning remains vulnerable to catastrophic forgetting, whose severity depends on how successive task gradients interact: when consecutive task gradients conflict, standard adapter initializations channel updates into subspaces that overwrite previously learned directions. We propose SLICE, a gradient-surgery-based initialization for LoRA adapters in continual learning. SLICE accumulates gradi"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Compared to vanilla LoRA, LoRA-GA, and LoRAM, SLICE consistently achieves a better stability-plasticity trade-off, improving Average Performance, Final Performance and Forgetting metrics while preserving General Performance and In Context Performance across both standard and adversarial continual learning sequences.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the projection operator applied to accumulated current-task and replay gradients, followed by truncated SVD, reliably channels updates into subspaces that avoid overwriting previously learned directions without introducing new interference or instability.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SLICE applies gradient surgery via projection and truncated SVD to initialize LoRA adapters, yielding better stability-plasticity trade-offs on continual learning benchmarks including adversarial task sequences.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SLICE initializes LoRA adapters by projecting current and replay gradients then applying truncated SVD to reduce catastrophic forgetting in continual learning.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"25e479a3484ec408eccf467ab27a62eba76eb507f14b17d49dc1527c5b4a1d49"},"source":{"id":"2605.12752","kind":"arxiv","version":1},"verdict":{"id":"b6def680-bc89-49a2-a35f-b05c0c8067ae","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:16:54.222262Z","strongest_claim":"Compared to vanilla LoRA, LoRA-GA, and LoRAM, SLICE consistently achieves a better stability-plasticity trade-off, improving Average Performance, Final Performance and Forgetting metrics while preserving General Performance and In Context Performance across both standard and adversarial continual learning sequences.","one_line_summary":"SLICE applies gradient surgery via projection and truncated SVD to initialize LoRA adapters, yielding better stability-plasticity trade-offs on continual learning benchmarks including adversarial task sequences.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the projection operator applied to accumulated current-task and replay gradients, followed by truncated SVD, reliably channels updates into subspaces that avoid overwriting previously learned directions without introducing new interference or instability.","pith_extraction_headline":"SLICE initializes LoRA adapters by projecting current and replay gradients then applying truncated SVD to reduce catastrophic forgetting in continual learning."},"references":{"count":41,"sample":[{"doi":"10.18653/v1/2022.emnlp-main.340","year":2022,"title":"and Hajishirzi, Hannaneh and Khashabi, Daniel , booktitle =","work_id":"2ff158b5-8900-4efa-8930-d26f842996ef","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"TRACE: A Comprehensive Benchmark for Continual Learning in Large Language Models , author=. 2023 , eprint=","work_id":"1fe4d920-7a63-40b6-adfb-a129c805c079","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=","work_id":"2e0a48e6-689d-4d5b-a9b3-e7d7556c8493","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Nature Machine Intelligence , volume=","work_id":"9a7cab73-2031-43c6-87c0-7448c501c3cb","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Chenlong Zhang and Zhuoran Jin and Hongbang Yuan and Jiaheng Wei and Tong Zhou and Kang Liu and Jun Zhao and Yubo Chen , booktitle=. 2025 , url=","work_id":"df6a0467-1c7e-45da-8b1d-08d657a32227","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":41,"snapshot_sha256":"0a9a3bfe92b1e52d14db27c9dff5b60844a0f7dbe5f82546b9643d6b4e64c321","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":"b6def680-bc89-49a2-a35f-b05c0c8067ae"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YCw3T2v7HaWSpICq69+j/HhKkTX4P7GVg+bnZ0HfRRMpkcMEfbmVG+mK0/PnmvlZ1x050M83O9CPn8ugrEYKBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T15:51:18.328099Z"},"content_sha256":"f5a1d5b743b6ae3fcaccba71dc03c4cc852e62ad42c1458334910807c07d0a19","schema_version":"1.0","event_id":"sha256:f5a1d5b743b6ae3fcaccba71dc03c4cc852e62ad42c1458334910807c07d0a19"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2MYR2HSV7CYX4VNAN26R6X3IYE/bundle.json","state_url":"https://pith.science/pith/2MYR2HSV7CYX4VNAN26R6X3IYE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2MYR2HSV7CYX4VNAN26R6X3IYE/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":"2026-05-29T15:51:18Z","links":{"resolver":"https://pith.science/pith/2MYR2HSV7CYX4VNAN26R6X3IYE","bundle":"https://pith.science/pith/2MYR2HSV7CYX4VNAN26R6X3IYE/bundle.json","state":"https://pith.science/pith/2MYR2HSV7CYX4VNAN26R6X3IYE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2MYR2HSV7CYX4VNAN26R6X3IYE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:2MYR2HSV7CYX4VNAN26R6X3IYE","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":"ece6f9960e995c19791123bd6a4d83755bd7db56e1331e3ac7c533225a36ecdb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T21:06:03Z","title_canon_sha256":"3e47e494325e05d37d6dd634f0e63dabbb1709d60463e7d92c67df390f757553"},"schema_version":"1.0","source":{"id":"2605.12752","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12752","created_at":"2026-05-18T03:09:48Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12752v1","created_at":"2026-05-18T03:09:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12752","created_at":"2026-05-18T03:09:48Z"},{"alias_kind":"pith_short_12","alias_value":"2MYR2HSV7CYX","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"2MYR2HSV7CYX4VNA","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"2MYR2HSV","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:f5a1d5b743b6ae3fcaccba71dc03c4cc852e62ad42c1458334910807c07d0a19","target":"graph","created_at":"2026-05-18T03:09:48Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Compared to vanilla LoRA, LoRA-GA, and LoRAM, SLICE consistently achieves a better stability-plasticity trade-off, improving Average Performance, Final Performance and Forgetting metrics while preserving General Performance and In Context Performance across both standard and adversarial continual learning sequences."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the projection operator applied to accumulated current-task and replay gradients, followed by truncated SVD, reliably channels updates into subspaces that avoid overwriting previously learned directions without introducing new interference or instability."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SLICE applies gradient surgery via projection and truncated SVD to initialize LoRA adapters, yielding better stability-plasticity trade-offs on continual learning benchmarks including adversarial task sequences."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"SLICE initializes LoRA adapters by projecting current and replay gradients then applying truncated SVD to reduce catastrophic forgetting in continual learning."}],"snapshot_sha256":"25e479a3484ec408eccf467ab27a62eba76eb507f14b17d49dc1527c5b4a1d49"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"LoRA is widely adopted for continual fine-tuning of Large Language Models due to its parameter efficiency, modularity across tasks, and compatibility with replay strategies. However, LoRA-based continual learning remains vulnerable to catastrophic forgetting, whose severity depends on how successive task gradients interact: when consecutive task gradients conflict, standard adapter initializations channel updates into subspaces that overwrite previously learned directions. We propose SLICE, a gradient-surgery-based initialization for LoRA adapters in continual learning. SLICE accumulates gradi","authors_text":"Arthur S. Bianchessi, Christian Mattjie, Joana Pasquali, Jo\\~ao Vitor Boer Abitante, Lucas S. Kupssinsk\\\"u, Ot\\'avio Parraga, Rafaela Cappelari Ravazio, Ramiro N. Barros, Rodrigo C. Barros, Vin\\'icius Conte Turani","cross_cats":[],"headline":"SLICE initializes LoRA adapters by projecting current and replay gradients then applying truncated SVD to reduce catastrophic forgetting in continual learning.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T21:06:03Z","title":"Low-Rank Adapters Initialization via Gradient Surgery for Continual Learning"},"references":{"count":41,"internal_anchors":0,"resolved_work":41,"sample":[{"cited_arxiv_id":"","doi":"10.18653/v1/2022.emnlp-main.340","is_internal_anchor":false,"ref_index":1,"title":"and Hajishirzi, Hannaneh and Khashabi, Daniel , booktitle =","work_id":"2ff158b5-8900-4efa-8930-d26f842996ef","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"TRACE: A Comprehensive Benchmark for Continual Learning in Large Language Models , author=. 2023 , eprint=","work_id":"1fe4d920-7a63-40b6-adfb-a129c805c079","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"The Thirty-ninth Annual Conference on Neural Information Processing Systems , year=","work_id":"2e0a48e6-689d-4d5b-a9b3-e7d7556c8493","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Nature Machine Intelligence , volume=","work_id":"9a7cab73-2031-43c6-87c0-7448c501c3cb","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Chenlong Zhang and Zhuoran Jin and Hongbang Yuan and Jiaheng Wei and Tong Zhou and Kang Liu and Jun Zhao and Yubo Chen , booktitle=. 2025 , url=","work_id":"df6a0467-1c7e-45da-8b1d-08d657a32227","year":2025}],"snapshot_sha256":"0a9a3bfe92b1e52d14db27c9dff5b60844a0f7dbe5f82546b9643d6b4e64c321"},"source":{"id":"2605.12752","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T21:16:54.222262Z","id":"b6def680-bc89-49a2-a35f-b05c0c8067ae","model_set":{"reader":"grok-4.3"},"one_line_summary":"SLICE applies gradient surgery via projection and truncated SVD to initialize LoRA adapters, yielding better stability-plasticity trade-offs on continual learning benchmarks including adversarial task sequences.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"SLICE initializes LoRA adapters by projecting current and replay gradients then applying truncated SVD to reduce catastrophic forgetting in continual learning.","strongest_claim":"Compared to vanilla LoRA, LoRA-GA, and LoRAM, SLICE consistently achieves a better stability-plasticity trade-off, improving Average Performance, Final Performance and Forgetting metrics while preserving General Performance and In Context Performance across both standard and adversarial continual learning sequences.","weakest_assumption":"That the projection operator applied to accumulated current-task and replay gradients, followed by truncated SVD, reliably channels updates into subspaces that avoid overwriting previously learned directions without introducing new interference or instability."}},"verdict_id":"b6def680-bc89-49a2-a35f-b05c0c8067ae"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:1f4d5393c009b3489df10822c5d77fcc39ff8cf56b2473312647557e67700235","target":"record","created_at":"2026-05-18T03:09:48Z","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":"ece6f9960e995c19791123bd6a4d83755bd7db56e1331e3ac7c533225a36ecdb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T21:06:03Z","title_canon_sha256":"3e47e494325e05d37d6dd634f0e63dabbb1709d60463e7d92c67df390f757553"},"schema_version":"1.0","source":{"id":"2605.12752","kind":"arxiv","version":1}},"canonical_sha256":"d3311d1e55f8b17e55a06ebd1f5f68c1025160fb359a75d8e4b80029c8ccd10d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d3311d1e55f8b17e55a06ebd1f5f68c1025160fb359a75d8e4b80029c8ccd10d","first_computed_at":"2026-05-18T03:09:48.735110Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:48.735110Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zQjeWlrb4LWuZ7/06TgyvrPNOkPm8FlLpmL1eAdH+JCe1fpg++vQP2KYfVCaWZrxyRNC0JEq7REhEYkiUZgfBA==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:48.735989Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12752","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1f4d5393c009b3489df10822c5d77fcc39ff8cf56b2473312647557e67700235","sha256:f5a1d5b743b6ae3fcaccba71dc03c4cc852e62ad42c1458334910807c07d0a19"],"state_sha256":"ac58b5d3ee477ab83ba4e9ea9b67aad1e33cd1d0d4cd7f6e41e493b8170b8b4f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vdcfckKOSePYU3svCl8ogdqmFffUEA1x0y54wpMfz2PUqLZEwglc828NgC74nG5YdkJ+KHlexk4Xyq4fvwK0AQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-29T15:51:18.333554Z","bundle_sha256":"d22cab122a3ef08929faf6ff120abe920173ea4ec38cba1afa082e696e1f9860"}}