{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:7IRT3RBDPNUWZFZUEDF2YDDBAT","short_pith_number":"pith:7IRT3RBD","canonical_record":{"source":{"id":"2605.12983","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T04:26:24Z","cross_cats_sorted":["cs.CC"],"title_canon_sha256":"3de8f5722741f44f17cbd7adeb96095d270d9b229d491aa0e77696099a562db7","abstract_canon_sha256":"5429ef58f1180b82d40d07b5d227c5f98126c71aebb2a3784a0f452c34fc980c"},"schema_version":"1.0"},"canonical_sha256":"fa233dc4237b696c973420cbac0c6104d7a3036322fc10e4a21b858b290e2fbb","source":{"kind":"arxiv","id":"2605.12983","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12983","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12983v1","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12983","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"pith_short_12","alias_value":"7IRT3RBDPNUW","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"7IRT3RBDPNUWZFZU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"7IRT3RBD","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:7IRT3RBDPNUWZFZUEDF2YDDBAT","target":"record","payload":{"canonical_record":{"source":{"id":"2605.12983","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T04:26:24Z","cross_cats_sorted":["cs.CC"],"title_canon_sha256":"3de8f5722741f44f17cbd7adeb96095d270d9b229d491aa0e77696099a562db7","abstract_canon_sha256":"5429ef58f1180b82d40d07b5d227c5f98126c71aebb2a3784a0f452c34fc980c"},"schema_version":"1.0"},"canonical_sha256":"fa233dc4237b696c973420cbac0c6104d7a3036322fc10e4a21b858b290e2fbb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:00.654196Z","signature_b64":"w1h4oftwlw7i+qPDq42f5H5Bmtz47CkITnggMZqXGe5khEIlkF5xGIAoRNAv3xkooEMozHNvWQASg3RQOvXNDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa233dc4237b696c973420cbac0c6104d7a3036322fc10e4a21b858b290e2fbb","last_reissued_at":"2026-05-18T03:09:00.653335Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:00.653335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.12983","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":"2026-05-18T03:09:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5B6HMbRSieorR+gzv6PEsWrgSJDLdwIq0u/gnumUlBRWKY/E68vnh9+uTVBBpn3b5UZqIYwseg2ozEUyf2wIBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T16:51:24.525783Z"},"content_sha256":"f4a343c16482c8aaee04ed9915daf96b7348e560f5ff40b4ad3e01a7c2cd4c65","schema_version":"1.0","event_id":"sha256:f4a343c16482c8aaee04ed9915daf96b7348e560f5ff40b4ad3e01a7c2cd4c65"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:7IRT3RBDPNUWZFZUEDF2YDDBAT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Decision Tree Learning on Product Spaces","license":"http://creativecommons.org/licenses/by/4.0/","headline":"For functions exactly computable by an optimal decision tree, the greedy heuristic produces an ε-approximating tree whose size is bounded by exp(Δ_opt D_opt log(e/ε)) under any product distribution.","cross_cats":["cs.CC"],"primary_cat":"cs.LG","authors_text":"Arshia Soltani Moakahr, Faraz Ghahremani, Kiarash Banihashem, MohammadTaghi Hajiaghayi","submitted_at":"2026-05-13T04:26:24Z","abstract_excerpt":"Decision tree learning has long been a central topic in theoretical computer science, driven by its practical importance. A fundamental and widely used method for decision tree construction is the top-down greedy heuristic, which recursively splits on the most influential variable. Despite its empirical success, theoretical analysis of this heuristic has been limited. A recent breakthrough by Blanc et al. (ITCS, 2020) provided the first rigorous theoretical guarantees for the greedy approach, but only under the uniform distribution. We extend this analysis to the more general and practically r"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our main result shows that for any function f computable by an optimal decision tree of size s, maximum depth D_opt, and average depth Δ_opt, the greedy heuristic constructs an ε-approximating tree whose size grows at most with exp(Δ_opt D_opt log(e/ε)).","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The analysis requires that the underlying distribution is a product distribution (features independent) and that the target function is exactly computable by some optimal decision tree with the stated size and depth parameters.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"For any function computable by an optimal decision tree with size s, max depth D_opt and average depth Δ_opt, the greedy heuristic builds an ε-approximating tree of size at most exp(Δ_opt D_opt log(e/ε)) under arbitrary product distributions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"For functions exactly computable by an optimal decision tree, the greedy heuristic produces an ε-approximating tree whose size is bounded by exp(Δ_opt D_opt log(e/ε)) under any product distribution.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"782f391b347fd14db33c6be24b83df129cac5dcbfda1b08dd8eae998aae7d3b1"},"source":{"id":"2605.12983","kind":"arxiv","version":1},"verdict":{"id":"0569217e-5927-42ab-86ab-53dac85cf389","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T20:06:14.894522Z","strongest_claim":"Our main result shows that for any function f computable by an optimal decision tree of size s, maximum depth D_opt, and average depth Δ_opt, the greedy heuristic constructs an ε-approximating tree whose size grows at most with exp(Δ_opt D_opt log(e/ε)).","one_line_summary":"For any function computable by an optimal decision tree with size s, max depth D_opt and average depth Δ_opt, the greedy heuristic builds an ε-approximating tree of size at most exp(Δ_opt D_opt log(e/ε)) under arbitrary product distributions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The analysis requires that the underlying distribution is a product distribution (features independent) and that the target function is exactly computable by some optimal decision tree with the stated size and depth parameters.","pith_extraction_headline":"For functions exactly computable by an optimal decision tree, the greedy heuristic produces an ε-approximating tree whose size is bounded by exp(Δ_opt D_opt log(e/ε)) under any product distribution."},"references":{"count":20,"sample":[{"doi":"10.4230/lipics.itcs.2020.44","year":2020,"title":"Top-Down Induction of Decision Trees: Rigorous Guarantees and Inherent Limitations , booktitle =","work_id":"4254b4dd-fbc1-4132-a41a-cda4fc8148e6","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"O'Donnell, R. and Saks, M. and Schramm, O. and Servedio, R.A. , booktitle=. Every decision tree has an influential variable , year=","work_id":"be558312-6cfd-48de-b464-a4a4609a2c96","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2002,"title":"Theoretical Computer Science , volume=","work_id":"97a984be-6a8a-40e5-99f3-8a2eabd95e06","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1989,"title":"Information and Computation , volume=","work_id":"6df16a39-ae01-46cd-b2a1-083c7fb18911","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Proceedings of the fortieth annual ACM symposium on Theory of computing , pages=","work_id":"c679f999-bc9a-452d-9d58-773be7b33d26","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":20,"snapshot_sha256":"2afe90f685c0723a9d6e230d2b464b9a9f290aab684f00c2269b4e3d729aff88","internal_anchors":1},"formal_canon":{"evidence_count":2,"snapshot_sha256":"be34e9e3aabc7262fe3e46465278a724bbc146b83a58b8664bd32482a79dc5c7"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"0569217e-5927-42ab-86ab-53dac85cf389"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ymy4WL68glpodK1yB8eq6HR45sDK+/Ildo+JDwH3GE324PTYJyaaj1toVLsbupwFOpePl8TjvaSprCO7YmlADQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T16:51:24.527058Z"},"content_sha256":"a36fc9d502d28a3f42f544490e67b7dbf7efb5b0e8805e067cf3009d06dee83c","schema_version":"1.0","event_id":"sha256:a36fc9d502d28a3f42f544490e67b7dbf7efb5b0e8805e067cf3009d06dee83c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7IRT3RBDPNUWZFZUEDF2YDDBAT/bundle.json","state_url":"https://pith.science/pith/7IRT3RBDPNUWZFZUEDF2YDDBAT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7IRT3RBDPNUWZFZUEDF2YDDBAT/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":"2026-05-29T16:51:24Z","links":{"resolver":"https://pith.science/pith/7IRT3RBDPNUWZFZUEDF2YDDBAT","bundle":"https://pith.science/pith/7IRT3RBDPNUWZFZUEDF2YDDBAT/bundle.json","state":"https://pith.science/pith/7IRT3RBDPNUWZFZUEDF2YDDBAT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7IRT3RBDPNUWZFZUEDF2YDDBAT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:7IRT3RBDPNUWZFZUEDF2YDDBAT","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":"5429ef58f1180b82d40d07b5d227c5f98126c71aebb2a3784a0f452c34fc980c","cross_cats_sorted":["cs.CC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T04:26:24Z","title_canon_sha256":"3de8f5722741f44f17cbd7adeb96095d270d9b229d491aa0e77696099a562db7"},"schema_version":"1.0","source":{"id":"2605.12983","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12983","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12983v1","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12983","created_at":"2026-05-18T03:09:00Z"},{"alias_kind":"pith_short_12","alias_value":"7IRT3RBDPNUW","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"7IRT3RBDPNUWZFZU","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"7IRT3RBD","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:a36fc9d502d28a3f42f544490e67b7dbf7efb5b0e8805e067cf3009d06dee83c","target":"graph","created_at":"2026-05-18T03:09:00Z","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 main result shows that for any function f computable by an optimal decision tree of size s, maximum depth D_opt, and average depth Δ_opt, the greedy heuristic constructs an ε-approximating tree whose size grows at most with exp(Δ_opt D_opt log(e/ε))."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The analysis requires that the underlying distribution is a product distribution (features independent) and that the target function is exactly computable by some optimal decision tree with the stated size and depth parameters."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"For any function computable by an optimal decision tree with size s, max depth D_opt and average depth Δ_opt, the greedy heuristic builds an ε-approximating tree of size at most exp(Δ_opt D_opt log(e/ε)) under arbitrary product distributions."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"For functions exactly computable by an optimal decision tree, the greedy heuristic produces an ε-approximating tree whose size is bounded by exp(Δ_opt D_opt log(e/ε)) under any product distribution."}],"snapshot_sha256":"782f391b347fd14db33c6be24b83df129cac5dcbfda1b08dd8eae998aae7d3b1"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"be34e9e3aabc7262fe3e46465278a724bbc146b83a58b8664bd32482a79dc5c7"},"paper":{"abstract_excerpt":"Decision tree learning has long been a central topic in theoretical computer science, driven by its practical importance. A fundamental and widely used method for decision tree construction is the top-down greedy heuristic, which recursively splits on the most influential variable. Despite its empirical success, theoretical analysis of this heuristic has been limited. A recent breakthrough by Blanc et al. (ITCS, 2020) provided the first rigorous theoretical guarantees for the greedy approach, but only under the uniform distribution. We extend this analysis to the more general and practically r","authors_text":"Arshia Soltani Moakahr, Faraz Ghahremani, Kiarash Banihashem, MohammadTaghi Hajiaghayi","cross_cats":["cs.CC"],"headline":"For functions exactly computable by an optimal decision tree, the greedy heuristic produces an ε-approximating tree whose size is bounded by exp(Δ_opt D_opt log(e/ε)) under any product distribution.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T04:26:24Z","title":"Decision Tree Learning on Product Spaces"},"references":{"count":20,"internal_anchors":1,"resolved_work":20,"sample":[{"cited_arxiv_id":"","doi":"10.4230/lipics.itcs.2020.44","is_internal_anchor":false,"ref_index":1,"title":"Top-Down Induction of Decision Trees: Rigorous Guarantees and Inherent Limitations , booktitle =","work_id":"4254b4dd-fbc1-4132-a41a-cda4fc8148e6","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"O'Donnell, R. and Saks, M. and Schramm, O. and Servedio, R.A. , booktitle=. Every decision tree has an influential variable , year=","work_id":"be558312-6cfd-48de-b464-a4a4609a2c96","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Theoretical Computer Science , volume=","work_id":"97a984be-6a8a-40e5-99f3-8a2eabd95e06","year":2002},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Information and Computation , volume=","work_id":"6df16a39-ae01-46cd-b2a1-083c7fb18911","year":1989},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Proceedings of the fortieth annual ACM symposium on Theory of computing , pages=","work_id":"c679f999-bc9a-452d-9d58-773be7b33d26","year":null}],"snapshot_sha256":"2afe90f685c0723a9d6e230d2b464b9a9f290aab684f00c2269b4e3d729aff88"},"source":{"id":"2605.12983","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T20:06:14.894522Z","id":"0569217e-5927-42ab-86ab-53dac85cf389","model_set":{"reader":"grok-4.3"},"one_line_summary":"For any function computable by an optimal decision tree with size s, max depth D_opt and average depth Δ_opt, the greedy heuristic builds an ε-approximating tree of size at most exp(Δ_opt D_opt log(e/ε)) under arbitrary product distributions.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"For functions exactly computable by an optimal decision tree, the greedy heuristic produces an ε-approximating tree whose size is bounded by exp(Δ_opt D_opt log(e/ε)) under any product distribution.","strongest_claim":"Our main result shows that for any function f computable by an optimal decision tree of size s, maximum depth D_opt, and average depth Δ_opt, the greedy heuristic constructs an ε-approximating tree whose size grows at most with exp(Δ_opt D_opt log(e/ε)).","weakest_assumption":"The analysis requires that the underlying distribution is a product distribution (features independent) and that the target function is exactly computable by some optimal decision tree with the stated size and depth parameters."}},"verdict_id":"0569217e-5927-42ab-86ab-53dac85cf389"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:f4a343c16482c8aaee04ed9915daf96b7348e560f5ff40b4ad3e01a7c2cd4c65","target":"record","created_at":"2026-05-18T03:09:00Z","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":"5429ef58f1180b82d40d07b5d227c5f98126c71aebb2a3784a0f452c34fc980c","cross_cats_sorted":["cs.CC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-13T04:26:24Z","title_canon_sha256":"3de8f5722741f44f17cbd7adeb96095d270d9b229d491aa0e77696099a562db7"},"schema_version":"1.0","source":{"id":"2605.12983","kind":"arxiv","version":1}},"canonical_sha256":"fa233dc4237b696c973420cbac0c6104d7a3036322fc10e4a21b858b290e2fbb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fa233dc4237b696c973420cbac0c6104d7a3036322fc10e4a21b858b290e2fbb","first_computed_at":"2026-05-18T03:09:00.653335Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:00.653335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"w1h4oftwlw7i+qPDq42f5H5Bmtz47CkITnggMZqXGe5khEIlkF5xGIAoRNAv3xkooEMozHNvWQASg3RQOvXNDA==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:00.654196Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12983","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f4a343c16482c8aaee04ed9915daf96b7348e560f5ff40b4ad3e01a7c2cd4c65","sha256:a36fc9d502d28a3f42f544490e67b7dbf7efb5b0e8805e067cf3009d06dee83c"],"state_sha256":"24c470747559a3b89b6f9e421cafe4199c7303784bcaa3ec5a529849f3605697"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9cK52Q9mmQu+EHLof+tF6EBVu6PAQwR7xhDkkYx/gtOXweAXuMuJtVaNWN9jWzlMGfVNdrp2Ax0fMH4Zn3iYDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-29T16:51:24.532785Z","bundle_sha256":"a096336777f047e693b9df54e78f21d69ab2ef59c7cfbe2ff32336c897ffbdab"}}