{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:KXHWQWJROEYUERR656X7RQJWGM","short_pith_number":"pith:KXHWQWJR","canonical_record":{"source":{"id":"2605.15368","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T19:47:43Z","cross_cats_sorted":["cs.GR","cs.LG"],"title_canon_sha256":"162b02cc3b151bf4c89920e7f54cde2d10b42dce221172b02e51fc07416ff144","abstract_canon_sha256":"c0f100e5e99f85d79c8d4149393bb5424a90de0bb430ab70f9e95d56f323c53d"},"schema_version":"1.0"},"canonical_sha256":"55cf685931713142463eefaff8c13633174cebb0bd6166335110dea30f2991e0","source":{"kind":"arxiv","id":"2605.15368","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15368","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15368v1","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15368","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"pith_short_12","alias_value":"KXHWQWJROEYU","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"pith_short_16","alias_value":"KXHWQWJROEYUERR6","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"pith_short_8","alias_value":"KXHWQWJR","created_at":"2026-05-20T00:00:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:KXHWQWJROEYUERR656X7RQJWGM","target":"record","payload":{"canonical_record":{"source":{"id":"2605.15368","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T19:47:43Z","cross_cats_sorted":["cs.GR","cs.LG"],"title_canon_sha256":"162b02cc3b151bf4c89920e7f54cde2d10b42dce221172b02e51fc07416ff144","abstract_canon_sha256":"c0f100e5e99f85d79c8d4149393bb5424a90de0bb430ab70f9e95d56f323c53d"},"schema_version":"1.0"},"canonical_sha256":"55cf685931713142463eefaff8c13633174cebb0bd6166335110dea30f2991e0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:00:54.889435Z","signature_b64":"wRgUDnEr9vfLDTJ6M4dYWa0hGoWK0kslkITe9TNBZBhJKHc7I5SgX0BAOZuqYe6gVy52A55Nk3IB7m1Na8lPCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55cf685931713142463eefaff8c13633174cebb0bd6166335110dea30f2991e0","last_reissued_at":"2026-05-20T00:00:54.888534Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:00:54.888534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.15368","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":"2026-05-20T00:00:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MBkSFJ3MAlHpBoptP4buaEsRIChbvGVjCQlKK8iHFZAFi98fwdW6Gfp5nlXEWcg2xjCfgua+CJ4zmNL+gsSfDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-20T15:52:33.079707Z"},"content_sha256":"7d9a9fdda117d357f9c6809c9c7b228fa6e17a2c83564e09feab21ddc85dd5f5","schema_version":"1.0","event_id":"sha256:7d9a9fdda117d357f9c6809c9c7b228fa6e17a2c83564e09feab21ddc85dd5f5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:KXHWQWJROEYUERR656X7RQJWGM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Discretizing Group-Convolutional Neural Networks for 3D Geometry in Feature Space","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Sampling transformations in feature space preserves accuracy in group-convolutional networks for 3D geometry while reducing costs.","cross_cats":["cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Daniel Franzen, Jean Philip Filling, Michael Wand","submitted_at":"2026-05-14T19:47:43Z","abstract_excerpt":"Group-convolutional neural networks (GCNNs) are among the most important methods for introducing symmetry as an inductive bias in deep learning: In each linear layer, GCNNs sample a transformation group $G$ densely and correlate data and filters in different poses (with suitable anti-aliasing for steerable GCNNs) to maintain equivariance with respect to $G$. Unfortunately, applying filters to many data items resulting from this sampling is expensive (even for translations alone, i.e., in ordinary CNNs), and costs grow exponentially with increasing degrees of freedom (such as translations and r"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"A coarse feature-space sampling already preserves classification accuracy remarkably well, which permits precomputation based on geometric similarity, accelerating the training of equivariant 3D classifiers substantially.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That representative samples selected purely by feature similarity are sufficient to maintain the equivariance properties and accuracy that dense geometric sampling would have provided.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Feature-space sampling in GCNNs preserves 3D classification accuracy with coarse discretization, enabling precomputation and faster training of equivariant models.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Sampling transformations in feature space preserves accuracy in group-convolutional networks for 3D geometry while reducing costs.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e9309d0f1ba48be07dd44cef7c2e52bf9863198263efa2b99aea10f406bc5e16"},"source":{"id":"2605.15368","kind":"arxiv","version":1},"verdict":{"id":"c117f56a-c84b-4460-a807-132dfab24720","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-19T15:52:59.929369Z","strongest_claim":"A coarse feature-space sampling already preserves classification accuracy remarkably well, which permits precomputation based on geometric similarity, accelerating the training of equivariant 3D classifiers substantially.","one_line_summary":"Feature-space sampling in GCNNs preserves 3D classification accuracy with coarse discretization, enabling precomputation and faster training of equivariant models.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That representative samples selected purely by feature similarity are sufficient to maintain the equivariance properties and accuracy that dense geometric sampling would have provided.","pith_extraction_headline":"Sampling transformations in feature space preserves accuracy in group-convolutional networks for 3D geometry while reducing costs."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.15368/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T16:04:43.779896Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T16:01:18.066954Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"claim_evidence","ran_at":"2026-05-19T14:21:54.188997Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.737802Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"0b29b75aa3de8283e814aea83285ca81baa2fc9945c511f4ed956dfeb9146c78"},"references":{"count":68,"sample":[{"doi":"","year":2018,"title":"Matan Atzmon, Haggai Maron, and Yaron Lipman. 2018. Point Convolutional Neural Networks by Extension Operators. Issue 4","work_id":"a80990c5-d780-4891-8b24-897a6cb24d52","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"Erik J. Bekkers. 2020. B-Spline CNNs on Lie Groups. InInternational Conference on Learning Representations (ICLR). https://openreview.net/forum?id=H1gBhkBFDH","work_id":"70d34c60-5913-43b4-9f1a-644ba707a4ed","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Bekkers, Sharvaree Vadgama, Rob D","work_id":"2ff25507-8729-43a9-be72-4ab69b6df00b","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Daniel Bolya, Cheng-Yang Fu, Xiaoliang Dai, Peizhao Zhang, Christoph Fe- ichtenhofer, and Judy Hoffman. 2023. Token Merging: Your ViT But Faster. InThe Eleventh International Conference on Learning Re","work_id":"e820c161-e107-4670-8cfd-b371dd00a16d","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Johann Brehmer, Sönke Behrends, Pim de Haan, and Taco Cohen. 2024. Does Equivariance Matter at Scale?arXiv preprint arXiv:2410.23179(2024). arXiv:2410.23179 [cs.LG]","work_id":"449c7cef-3264-4ca5-8b94-5558faef1bc7","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":68,"snapshot_sha256":"c00be62b5c0c94b69725b84e3a3aa7149a608e92c184ec6578429df3180199db","internal_anchors":3},"formal_canon":{"evidence_count":1,"snapshot_sha256":"17daebc31e25b10a0ec96f89c543c9bf853023f7f6df6369aee0196b2cd76358"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"c117f56a-c84b-4460-a807-132dfab24720"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:00:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PiN0mkdzjJtAosQA+XdEq1ADq0vxCavqYKIzDualiEvartfIb4F2ocOgYfxa4bHCvnnm4vyXvPYuergW5W2XCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-20T15:52:33.080369Z"},"content_sha256":"c724565450b43fc09a46d78074f2eb03c5c3808dd1f7c0192de0c00c43387c25","schema_version":"1.0","event_id":"sha256:c724565450b43fc09a46d78074f2eb03c5c3808dd1f7c0192de0c00c43387c25"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KXHWQWJROEYUERR656X7RQJWGM/bundle.json","state_url":"https://pith.science/pith/KXHWQWJROEYUERR656X7RQJWGM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KXHWQWJROEYUERR656X7RQJWGM/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":"2026-05-20T15:52:33Z","links":{"resolver":"https://pith.science/pith/KXHWQWJROEYUERR656X7RQJWGM","bundle":"https://pith.science/pith/KXHWQWJROEYUERR656X7RQJWGM/bundle.json","state":"https://pith.science/pith/KXHWQWJROEYUERR656X7RQJWGM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KXHWQWJROEYUERR656X7RQJWGM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:KXHWQWJROEYUERR656X7RQJWGM","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":"c0f100e5e99f85d79c8d4149393bb5424a90de0bb430ab70f9e95d56f323c53d","cross_cats_sorted":["cs.GR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T19:47:43Z","title_canon_sha256":"162b02cc3b151bf4c89920e7f54cde2d10b42dce221172b02e51fc07416ff144"},"schema_version":"1.0","source":{"id":"2605.15368","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.15368","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"arxiv_version","alias_value":"2605.15368v1","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.15368","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"pith_short_12","alias_value":"KXHWQWJROEYU","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"pith_short_16","alias_value":"KXHWQWJROEYUERR6","created_at":"2026-05-20T00:00:54Z"},{"alias_kind":"pith_short_8","alias_value":"KXHWQWJR","created_at":"2026-05-20T00:00:54Z"}],"graph_snapshots":[{"event_id":"sha256:c724565450b43fc09a46d78074f2eb03c5c3808dd1f7c0192de0c00c43387c25","target":"graph","created_at":"2026-05-20T00:00:54Z","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":"A coarse feature-space sampling already preserves classification accuracy remarkably well, which permits precomputation based on geometric similarity, accelerating the training of equivariant 3D classifiers substantially."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That representative samples selected purely by feature similarity are sufficient to maintain the equivariance properties and accuracy that dense geometric sampling would have provided."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Feature-space sampling in GCNNs preserves 3D classification accuracy with coarse discretization, enabling precomputation and faster training of equivariant models."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Sampling transformations in feature space preserves accuracy in group-convolutional networks for 3D geometry while reducing costs."}],"snapshot_sha256":"e9309d0f1ba48be07dd44cef7c2e52bf9863198263efa2b99aea10f406bc5e16"},"formal_canon":{"evidence_count":1,"snapshot_sha256":"17daebc31e25b10a0ec96f89c543c9bf853023f7f6df6369aee0196b2cd76358"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T16:04:43.779896Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T16:01:18.066954Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T14:21:54.188997Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:22.737802Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.15368/integrity.json","findings":[],"snapshot_sha256":"0b29b75aa3de8283e814aea83285ca81baa2fc9945c511f4ed956dfeb9146c78","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Group-convolutional neural networks (GCNNs) are among the most important methods for introducing symmetry as an inductive bias in deep learning: In each linear layer, GCNNs sample a transformation group $G$ densely and correlate data and filters in different poses (with suitable anti-aliasing for steerable GCNNs) to maintain equivariance with respect to $G$. Unfortunately, applying filters to many data items resulting from this sampling is expensive (even for translations alone, i.e., in ordinary CNNs), and costs grow exponentially with increasing degrees of freedom (such as translations and r","authors_text":"Daniel Franzen, Jean Philip Filling, Michael Wand","cross_cats":["cs.GR","cs.LG"],"headline":"Sampling transformations in feature space preserves accuracy in group-convolutional networks for 3D geometry while reducing costs.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T19:47:43Z","title":"Discretizing Group-Convolutional Neural Networks for 3D Geometry in Feature Space"},"references":{"count":68,"internal_anchors":3,"resolved_work":68,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Matan Atzmon, Haggai Maron, and Yaron Lipman. 2018. Point Convolutional Neural Networks by Extension Operators. Issue 4","work_id":"a80990c5-d780-4891-8b24-897a6cb24d52","year":2018},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Erik J. Bekkers. 2020. B-Spline CNNs on Lie Groups. InInternational Conference on Learning Representations (ICLR). https://openreview.net/forum?id=H1gBhkBFDH","work_id":"70d34c60-5913-43b4-9f1a-644ba707a4ed","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Bekkers, Sharvaree Vadgama, Rob D","work_id":"2ff25507-8729-43a9-be72-4ab69b6df00b","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Daniel Bolya, Cheng-Yang Fu, Xiaoliang Dai, Peizhao Zhang, Christoph Fe- ichtenhofer, and Judy Hoffman. 2023. Token Merging: Your ViT But Faster. InThe Eleventh International Conference on Learning Re","work_id":"e820c161-e107-4670-8cfd-b371dd00a16d","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Johann Brehmer, Sönke Behrends, Pim de Haan, and Taco Cohen. 2024. Does Equivariance Matter at Scale?arXiv preprint arXiv:2410.23179(2024). arXiv:2410.23179 [cs.LG]","work_id":"449c7cef-3264-4ca5-8b94-5558faef1bc7","year":2024}],"snapshot_sha256":"c00be62b5c0c94b69725b84e3a3aa7149a608e92c184ec6578429df3180199db"},"source":{"id":"2605.15368","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T15:52:59.929369Z","id":"c117f56a-c84b-4460-a807-132dfab24720","model_set":{"reader":"grok-4.3"},"one_line_summary":"Feature-space sampling in GCNNs preserves 3D classification accuracy with coarse discretization, enabling precomputation and faster training of equivariant models.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Sampling transformations in feature space preserves accuracy in group-convolutional networks for 3D geometry while reducing costs.","strongest_claim":"A coarse feature-space sampling already preserves classification accuracy remarkably well, which permits precomputation based on geometric similarity, accelerating the training of equivariant 3D classifiers substantially.","weakest_assumption":"That representative samples selected purely by feature similarity are sufficient to maintain the equivariance properties and accuracy that dense geometric sampling would have provided."}},"verdict_id":"c117f56a-c84b-4460-a807-132dfab24720"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:7d9a9fdda117d357f9c6809c9c7b228fa6e17a2c83564e09feab21ddc85dd5f5","target":"record","created_at":"2026-05-20T00:00:54Z","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":"c0f100e5e99f85d79c8d4149393bb5424a90de0bb430ab70f9e95d56f323c53d","cross_cats_sorted":["cs.GR","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T19:47:43Z","title_canon_sha256":"162b02cc3b151bf4c89920e7f54cde2d10b42dce221172b02e51fc07416ff144"},"schema_version":"1.0","source":{"id":"2605.15368","kind":"arxiv","version":1}},"canonical_sha256":"55cf685931713142463eefaff8c13633174cebb0bd6166335110dea30f2991e0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"55cf685931713142463eefaff8c13633174cebb0bd6166335110dea30f2991e0","first_computed_at":"2026-05-20T00:00:54.888534Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:00:54.888534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wRgUDnEr9vfLDTJ6M4dYWa0hGoWK0kslkITe9TNBZBhJKHc7I5SgX0BAOZuqYe6gVy52A55Nk3IB7m1Na8lPCQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:00:54.889435Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.15368","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7d9a9fdda117d357f9c6809c9c7b228fa6e17a2c83564e09feab21ddc85dd5f5","sha256:c724565450b43fc09a46d78074f2eb03c5c3808dd1f7c0192de0c00c43387c25"],"state_sha256":"5d9105143fd6ac137b982c655cc06224717c75d468096cd088da75eeabf2c12c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7PQ27EFB5qss+LMxaRTn3M8iNQjaEDL9llFAslYI8hL4CophfbflY7oC0HwOsQ6t/LsWg6P1agbk1WVWmvTEDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-20T15:52:33.083010Z","bundle_sha256":"75160e1b506a01e0837e0d314824f124f5da34758773f9f58df02ea2d7530588"}}