{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:YV5CIYVKXAEKEDRDVCDIWAO5LA","short_pith_number":"pith:YV5CIYVK","canonical_record":{"source":{"id":"1809.11096","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-09-28T15:38:49Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"8a91f757ab4b07b80ade8b31df7f9a2b2c14c63fdb422e193f09c9bce7e2138d","abstract_canon_sha256":"0fe46bc3c1d55080236dd18c68dd0d3d84e2a254652892aebd6e51a8ab824be9"},"schema_version":"1.0"},"canonical_sha256":"c57a2462aab808a20e23a8868b01dd580571b1f9f06b8fd2527a25394dac2aaf","source":{"kind":"arxiv","id":"1809.11096","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1809.11096","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"arxiv_version","alias_value":"1809.11096v2","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.11096","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"pith_short_12","alias_value":"YV5CIYVKXAEK","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"pith_short_16","alias_value":"YV5CIYVKXAEKEDRD","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"pith_short_8","alias_value":"YV5CIYVK","created_at":"2026-07-04T23:25:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:YV5CIYVKXAEKEDRDVCDIWAO5LA","target":"record","payload":{"canonical_record":{"source":{"id":"1809.11096","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-09-28T15:38:49Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"8a91f757ab4b07b80ade8b31df7f9a2b2c14c63fdb422e193f09c9bce7e2138d","abstract_canon_sha256":"0fe46bc3c1d55080236dd18c68dd0d3d84e2a254652892aebd6e51a8ab824be9"},"schema_version":"1.0"},"canonical_sha256":"c57a2462aab808a20e23a8868b01dd580571b1f9f06b8fd2527a25394dac2aaf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:25:17.407000Z","signature_b64":"8NvjF9fvs52h16PT+jlfMYzAOwV5F4XPG2FRNHPYXP8BPhOGhd6Ry/ezd4hWowpGFeXkl+J1UzkYN7rx5AhICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c57a2462aab808a20e23a8868b01dd580571b1f9f06b8fd2527a25394dac2aaf","last_reissued_at":"2026-07-04T23:25:17.406489Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:25:17.406489Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1809.11096","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":"2026-07-04T23:25:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SamKrP3gMwRF7SaFyIRQ/iYTnlm0VLRnvrfEHkDdVdrXzioCSfby+T3kcFFdFPdD8m877S4n+PdvJEvggfbyCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T03:34:56.757248Z"},"content_sha256":"305ba13c632b90072a398d9bc7057ca3d0fb661809f7eb2fecc4a50300f930a1","schema_version":"1.0","event_id":"sha256:305ba13c632b90072a398d9bc7057ca3d0fb661809f7eb2fecc4a50300f930a1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:YV5CIYVKXAEKEDRDVCDIWAO5LA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Large Scale GAN Training for High Fidelity Natural Image Synthesis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Scaling GAN training with orthogonal regularization and a truncation trick yields new state-of-the-art results on class-conditional ImageNet synthesis.","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Brock, Jeff Donahue, Karen Simonyan","submitted_at":"2018-09-28T15:38:49Z","abstract_excerpt":"Despite recent progress in generative image modeling, successfully generating high-resolution, diverse samples from complex datasets such as ImageNet remains an elusive goal. To this end, we train Generative Adversarial Networks at the largest scale yet attempted, and study the instabilities specific to such scale. We find that applying orthogonal regularization to the generator renders it amenable to a simple \"truncation trick,\" allowing fine control over the trade-off between sample fidelity and variety by reducing the variance of the Generator's input. Our modifications lead to models which"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our modifications lead to models which set the new state of the art in class-conditional image synthesis. When trained on ImageNet at 128x128 resolution, our models (BigGANs) achieve an Inception Score (IS) of 166.5 and Frechet Inception Distance (FID) of 7.4, improving over the previous best IS of 52.52 and FID of 18.6.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the reported IS and FID metrics, measured on standard splits, reliably indicate human-perceived fidelity and that the orthogonal regularization plus truncation will generalize without hidden instabilities or mode collapse on other datasets or resolutions.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"BigGANs achieve state-of-the-art class-conditional synthesis on ImageNet 128x128 with Inception Score 166.5 and FID 7.4 by scaling GANs and applying orthogonal regularization plus truncation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Scaling GAN training with orthogonal regularization and a truncation trick yields new state-of-the-art results on class-conditional ImageNet synthesis.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e4d0eb79323422e14d7ddb089cd84e522269417693d318bbac28b31a2cfb2966"},"source":{"id":"1809.11096","kind":"arxiv","version":2},"verdict":{"id":"14e29915-59dc-41ae-8646-96a849d38f49","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T12:07:01.086101Z","strongest_claim":"Our modifications lead to models which set the new state of the art in class-conditional image synthesis. When trained on ImageNet at 128x128 resolution, our models (BigGANs) achieve an Inception Score (IS) of 166.5 and Frechet Inception Distance (FID) of 7.4, improving over the previous best IS of 52.52 and FID of 18.6.","one_line_summary":"BigGANs achieve state-of-the-art class-conditional synthesis on ImageNet 128x128 with Inception Score 166.5 and FID 7.4 by scaling GANs and applying orthogonal regularization plus truncation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the reported IS and FID metrics, measured on standard splits, reliably indicate human-perceived fidelity and that the orthogonal regularization plus truncation will generalize without hidden instabilities or mode collapse on other datasets or resolutions.","pith_extraction_headline":"Scaling GAN training with orthogonal regularization and a truncation trick yields new state-of-the-art results on class-conditional ImageNet synthesis."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1809.11096/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":34,"sample":[{"doi":"","year":null,"title":"A Note on the Inception Score","work_id":"4700104e-68bd-4452-bdd6-e68281808831","ref_index":1,"cited_arxiv_id":"1801.01973","is_internal_anchor":false},{"doi":"","year":null,"title":"The Cramer Distance as a Solution to Biased Wasserstein Gradients","work_id":"5dfdb668-ff19-4c96-acc8-bc352a3ec469","ref_index":2,"cited_arxiv_id":"1705.10743","is_internal_anchor":false},{"doi":"","year":2019,"title":"Dai, Shakir Mohamed, and Ian Goodfellow","work_id":"e7795d41-ff9d-43d6-843e-15740654f868","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"On Convergence and Stability of GANs","work_id":"55d348e9-cc9f-483b-99d9-5a601a58eb21","ref_index":4,"cited_arxiv_id":"1705.07215","is_internal_anchor":false},{"doi":"","year":null,"title":"Geometric GAN","work_id":"a33c5863-4b4e-4797-8479-3fca1344dccd","ref_index":5,"cited_arxiv_id":"1705.02894","is_internal_anchor":false}],"resolved_work":34,"snapshot_sha256":"0d8f54e4c8d1dd88de0ba520dc6cd70a8d0737707aa8bc6de72047bd64ed24b6","internal_anchors":4},"formal_canon":{"evidence_count":2,"snapshot_sha256":"464a511c6515f700c4f52690d6f096f630bceffa1d3eb6bd500f0e5ff726379a"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"14e29915-59dc-41ae-8646-96a849d38f49"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-04T23:25:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"s7+XiFpmfW3UhogCdsgwZv+Pi/QR2o//Z5IOR1BC2jqprPzdoPzF1bMu4+PqrCI1os94VTM6wk+ApDgGK36EBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T03:34:56.758183Z"},"content_sha256":"c35a3d4d2c0e8835652ffbaf8a90aa8ee3eb6001882bae66e60ef3a45d2d84e4","schema_version":"1.0","event_id":"sha256:c35a3d4d2c0e8835652ffbaf8a90aa8ee3eb6001882bae66e60ef3a45d2d84e4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YV5CIYVKXAEKEDRDVCDIWAO5LA/bundle.json","state_url":"https://pith.science/pith/YV5CIYVKXAEKEDRDVCDIWAO5LA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YV5CIYVKXAEKEDRDVCDIWAO5LA/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":"2026-08-12T03:34:56Z","links":{"resolver":"https://pith.science/pith/YV5CIYVKXAEKEDRDVCDIWAO5LA","bundle":"https://pith.science/pith/YV5CIYVKXAEKEDRDVCDIWAO5LA/bundle.json","state":"https://pith.science/pith/YV5CIYVKXAEKEDRDVCDIWAO5LA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YV5CIYVKXAEKEDRDVCDIWAO5LA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:YV5CIYVKXAEKEDRDVCDIWAO5LA","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":"0fe46bc3c1d55080236dd18c68dd0d3d84e2a254652892aebd6e51a8ab824be9","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-09-28T15:38:49Z","title_canon_sha256":"8a91f757ab4b07b80ade8b31df7f9a2b2c14c63fdb422e193f09c9bce7e2138d"},"schema_version":"1.0","source":{"id":"1809.11096","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1809.11096","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"arxiv_version","alias_value":"1809.11096v2","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.11096","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"pith_short_12","alias_value":"YV5CIYVKXAEK","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"pith_short_16","alias_value":"YV5CIYVKXAEKEDRD","created_at":"2026-07-04T23:25:17Z"},{"alias_kind":"pith_short_8","alias_value":"YV5CIYVK","created_at":"2026-07-04T23:25:17Z"}],"graph_snapshots":[{"event_id":"sha256:c35a3d4d2c0e8835652ffbaf8a90aa8ee3eb6001882bae66e60ef3a45d2d84e4","target":"graph","created_at":"2026-07-04T23:25:17Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Our modifications lead to models which set the new state of the art in class-conditional image synthesis. When trained on ImageNet at 128x128 resolution, our models (BigGANs) achieve an Inception Score (IS) of 166.5 and Frechet Inception Distance (FID) of 7.4, improving over the previous best IS of 52.52 and FID of 18.6."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the reported IS and FID metrics, measured on standard splits, reliably indicate human-perceived fidelity and that the orthogonal regularization plus truncation will generalize without hidden instabilities or mode collapse on other datasets or resolutions."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"BigGANs achieve state-of-the-art class-conditional synthesis on ImageNet 128x128 with Inception Score 166.5 and FID 7.4 by scaling GANs and applying orthogonal regularization plus truncation."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Scaling GAN training with orthogonal regularization and a truncation trick yields new state-of-the-art results on class-conditional ImageNet synthesis."}],"snapshot_sha256":"e4d0eb79323422e14d7ddb089cd84e522269417693d318bbac28b31a2cfb2966"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"464a511c6515f700c4f52690d6f096f630bceffa1d3eb6bd500f0e5ff726379a"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/1809.11096/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite recent progress in generative image modeling, successfully generating high-resolution, diverse samples from complex datasets such as ImageNet remains an elusive goal. To this end, we train Generative Adversarial Networks at the largest scale yet attempted, and study the instabilities specific to such scale. We find that applying orthogonal regularization to the generator renders it amenable to a simple \"truncation trick,\" allowing fine control over the trade-off between sample fidelity and variety by reducing the variance of the Generator's input. Our modifications lead to models which","authors_text":"Andrew Brock, Jeff Donahue, Karen Simonyan","cross_cats":["stat.ML"],"headline":"Scaling GAN training with orthogonal regularization and a truncation trick yields new state-of-the-art results on class-conditional ImageNet synthesis.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-09-28T15:38:49Z","title":"Large Scale GAN Training for High Fidelity Natural Image Synthesis"},"references":{"count":34,"internal_anchors":4,"resolved_work":34,"sample":[{"cited_arxiv_id":"1801.01973","doi":"","is_internal_anchor":false,"ref_index":1,"title":"A Note on the Inception Score","work_id":"4700104e-68bd-4452-bdd6-e68281808831","year":null},{"cited_arxiv_id":"1705.10743","doi":"","is_internal_anchor":false,"ref_index":2,"title":"The Cramer Distance as a Solution to Biased Wasserstein Gradients","work_id":"5dfdb668-ff19-4c96-acc8-bc352a3ec469","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Dai, Shakir Mohamed, and Ian Goodfellow","work_id":"e7795d41-ff9d-43d6-843e-15740654f868","year":2019},{"cited_arxiv_id":"1705.07215","doi":"","is_internal_anchor":false,"ref_index":4,"title":"On Convergence and Stability of GANs","work_id":"55d348e9-cc9f-483b-99d9-5a601a58eb21","year":null},{"cited_arxiv_id":"1705.02894","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Geometric GAN","work_id":"a33c5863-4b4e-4797-8479-3fca1344dccd","year":null}],"snapshot_sha256":"0d8f54e4c8d1dd88de0ba520dc6cd70a8d0737707aa8bc6de72047bd64ed24b6"},"source":{"id":"1809.11096","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-13T12:07:01.086101Z","id":"14e29915-59dc-41ae-8646-96a849d38f49","model_set":{"reader":"grok-4.3"},"one_line_summary":"BigGANs achieve state-of-the-art class-conditional synthesis on ImageNet 128x128 with Inception Score 166.5 and FID 7.4 by scaling GANs and applying orthogonal regularization plus truncation.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Scaling GAN training with orthogonal regularization and a truncation trick yields new state-of-the-art results on class-conditional ImageNet synthesis.","strongest_claim":"Our modifications lead to models which set the new state of the art in class-conditional image synthesis. When trained on ImageNet at 128x128 resolution, our models (BigGANs) achieve an Inception Score (IS) of 166.5 and Frechet Inception Distance (FID) of 7.4, improving over the previous best IS of 52.52 and FID of 18.6.","weakest_assumption":"That the reported IS and FID metrics, measured on standard splits, reliably indicate human-perceived fidelity and that the orthogonal regularization plus truncation will generalize without hidden instabilities or mode collapse on other datasets or resolutions."}},"verdict_id":"14e29915-59dc-41ae-8646-96a849d38f49"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:305ba13c632b90072a398d9bc7057ca3d0fb661809f7eb2fecc4a50300f930a1","target":"record","created_at":"2026-07-04T23:25:17Z","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":"0fe46bc3c1d55080236dd18c68dd0d3d84e2a254652892aebd6e51a8ab824be9","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-09-28T15:38:49Z","title_canon_sha256":"8a91f757ab4b07b80ade8b31df7f9a2b2c14c63fdb422e193f09c9bce7e2138d"},"schema_version":"1.0","source":{"id":"1809.11096","kind":"arxiv","version":2}},"canonical_sha256":"c57a2462aab808a20e23a8868b01dd580571b1f9f06b8fd2527a25394dac2aaf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c57a2462aab808a20e23a8868b01dd580571b1f9f06b8fd2527a25394dac2aaf","first_computed_at":"2026-07-04T23:25:17.406489Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-04T23:25:17.406489Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8NvjF9fvs52h16PT+jlfMYzAOwV5F4XPG2FRNHPYXP8BPhOGhd6Ry/ezd4hWowpGFeXkl+J1UzkYN7rx5AhICg==","signature_status":"signed_v1","signed_at":"2026-07-04T23:25:17.407000Z","signed_message":"canonical_sha256_bytes"},"source_id":"1809.11096","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:305ba13c632b90072a398d9bc7057ca3d0fb661809f7eb2fecc4a50300f930a1","sha256:c35a3d4d2c0e8835652ffbaf8a90aa8ee3eb6001882bae66e60ef3a45d2d84e4"],"state_sha256":"801dfecf0f472b0feb8542be50d5e447add777d13a570aa6786dfd3bf1ab274d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ukLvHXZUevHVozBCIFI+B/8uxKBNEEAeYJiPdgN9Rl4NvcFoUNOgghvrD75kUWuXiNUowufcMsyWvyKF+9MIBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T03:34:56.764934Z","bundle_sha256":"6f36d0c98f050e0b1cfe42e3b132881776204e2275f3f30efbeb033b0fdee120"}}