{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:MV3FB4SECH3OERERHGKKY3XGRW","short_pith_number":"pith:MV3FB4SE","canonical_record":{"source":{"id":"2605.14141","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T21:47:55Z","cross_cats_sorted":[],"title_canon_sha256":"a4a5f43679d3e94f6e06361665ca378d546a6fe1fd264a3934590f9378a3d406","abstract_canon_sha256":"0c338b5ee72081852eae22d591daca02331ef2758aa81e032626ff43fc4b8bf7"},"schema_version":"1.0"},"canonical_sha256":"657650f24411f6e244913994ac6ee68da9d899b2b40242601aafbb71ee298f33","source":{"kind":"arxiv","id":"2605.14141","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14141","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14141v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14141","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"MV3FB4SECH3O","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"MV3FB4SECH3OERER","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"MV3FB4SE","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:MV3FB4SECH3OERERHGKKY3XGRW","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14141","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T21:47:55Z","cross_cats_sorted":[],"title_canon_sha256":"a4a5f43679d3e94f6e06361665ca378d546a6fe1fd264a3934590f9378a3d406","abstract_canon_sha256":"0c338b5ee72081852eae22d591daca02331ef2758aa81e032626ff43fc4b8bf7"},"schema_version":"1.0"},"canonical_sha256":"657650f24411f6e244913994ac6ee68da9d899b2b40242601aafbb71ee298f33","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:11.688177Z","signature_b64":"jCxdTz46R6DeRwNYrTszg1RFLB4MtdJEOq/qh1mZiSyGeugbjrBQXSBV8HPUKjDw+t5mmbIYnd5n62rRMQuiDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"657650f24411f6e244913994ac6ee68da9d899b2b40242601aafbb71ee298f33","last_reissued_at":"2026-05-17T23:39:11.687677Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:11.687677Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14141","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":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GTAPsYVM1r3JGpb1zdq2E2pfe4MbZw9rBUUAdVhsKOwQbW4gn20T2mDlxmPYdRsyqA2MWXVv9eLA3f0r6KfPCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T06:08:35.301070Z"},"content_sha256":"ba7aef28efd9f0ad897809c23363b2360b20bba735df793bd96bd0a27d2c7f50","schema_version":"1.0","event_id":"sha256:ba7aef28efd9f0ad897809c23363b2360b20bba735df793bd96bd0a27d2c7f50"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:MV3FB4SECH3OERERHGKKY3XGRW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Distribution-Aware Algorithm Design with LLM Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"The empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime to new instances from the same distribution.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Pierfrancesco Beneventano, Priyadarsi Mishra, Saharsh Koganti, Tomer Galanti","submitted_at":"2026-05-13T21:47:55Z","abstract_excerpt":"We study learning when the learned object is executable solver code rather than a predictor. In this setting, correctness is not enough: two solvers may both return valid solutions on the deployment distribution while differing substantially in runtime. Given samples from an unknown task distribution, the learner returns code evaluated on fresh instances by both solution quality and execution time. Our central abstraction is a \\emph{solver hint}: reusable structure inferred from samples and compiled into specialized solver code. We prove that the empirically fastest sample-consistent solver fr"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We prove that the empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime, and that statistically identifiable hints can be recovered and compiled from polynomially many samples.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That statistically identifiable solver hints exist for the target distributions and can be recovered from polynomially many samples to produce a fixed library of solvers whose fastest member generalizes in runtime.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LLM agents synthesize distribution-aware solver code that achieves 0.971 mean normalized quality and runs 336x faster than top heuristics on 21 combinatorial optimization distributions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"The empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime to new instances from the same distribution.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"f36db2151c12d8105bdff8d72cf2e1e215409e81b7a5bd921e7ba5a0343a080d"},"source":{"id":"2605.14141","kind":"arxiv","version":1},"verdict":{"id":"d0c02064-3723-4809-a565-2fae46ef4866","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T04:59:58.226238Z","strongest_claim":"We prove that the empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime, and that statistically identifiable hints can be recovered and compiled from polynomially many samples.","one_line_summary":"LLM agents synthesize distribution-aware solver code that achieves 0.971 mean normalized quality and runs 336x faster than top heuristics on 21 combinatorial optimization distributions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That statistically identifiable solver hints exist for the target distributions and can be recovered from polynomially many samples to produce a fixed library of solvers whose fastest member generalizes in runtime.","pith_extraction_headline":"The empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime to new instances from the same distribution."},"references":{"count":59,"sample":[{"doi":"","year":2006,"title":"D. L. Applegate, R. E. Bixby, V . Chvátal, and W. J. Cook.The Traveling Salesman Problem: A Computational Study. Princeton Series in Applied Mathematics. Princeton University Press, 2006","work_id":"da3eb0be-b01d-4827-9ac8-1747e399b9ec","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2020,"title":"F. Avellaneda. A short description of the solver EvalMaxSAT. In F. Bacchus, J. Berg, M. Järvisalo, and R. Martins, editors,MaxSAT Evaluation 2020: Solver and Benchmark Descriptions, pages 8–9, 2020","work_id":"d1c88265-eefa-4067-8276-93409ebadf16","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"M. Balog, A. L. Gaunt, M. Brockschmidt, S. Nowozin, and D. Tarlow. Deepcoder: Learning to write programs. InInternational Conference on Learning Representations, 2017","work_id":"4c4dd0a9-5ebd-49be-8f9a-1b1ca63949bd","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"K. Bestuzheva, M. Besançon, W.-K. Chen, A. Chmiela, T. Donkiewicz, J. van Doornmalen, L. Eifler, O. Gaul, G. Gamrath, A. Gleixner, et al. Enabling research through the SCIP optimization suite 8.0.ACM ","work_id":"7db8e0b5-2232-4718-bc8f-e7e5427ffaf3","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2006,"title":"A. Bogdanov and L. Trevisan. Average-case complexity.Foundations and Trends in Theoretical Computer Science, 2(1):1–106, 2006","work_id":"14263d28-0c99-4063-afd5-4a3b0edfb8c0","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":59,"snapshot_sha256":"2f77fb2e5f7a114b88cab5aed36dcd24b82c48243e13f7f108993c4939ce8a42","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"982a0fab1c9f213162ee85cd2b8d0eb4df7f4973992311d9a18ad5c82d720aa3"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"d0c02064-3723-4809-a565-2fae46ef4866"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sSaT6z9NDkxepCxcGCqStFlxrwA1BZ9ozPuXL/GKUYTwja3x9JyQF0Gj4m7yQwnogoo99bo6KGij12XUTkxaAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-02T06:08:35.301756Z"},"content_sha256":"db418f933981ff32be1140e2e4ed5981897c7630d04f4eacb90c9f2a046308fc","schema_version":"1.0","event_id":"sha256:db418f933981ff32be1140e2e4ed5981897c7630d04f4eacb90c9f2a046308fc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MV3FB4SECH3OERERHGKKY3XGRW/bundle.json","state_url":"https://pith.science/pith/MV3FB4SECH3OERERHGKKY3XGRW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MV3FB4SECH3OERERHGKKY3XGRW/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":"2026-06-02T06:08:35Z","links":{"resolver":"https://pith.science/pith/MV3FB4SECH3OERERHGKKY3XGRW","bundle":"https://pith.science/pith/MV3FB4SECH3OERERHGKKY3XGRW/bundle.json","state":"https://pith.science/pith/MV3FB4SECH3OERERHGKKY3XGRW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MV3FB4SECH3OERERHGKKY3XGRW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MV3FB4SECH3OERERHGKKY3XGRW","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":"0c338b5ee72081852eae22d591daca02331ef2758aa81e032626ff43fc4b8bf7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T21:47:55Z","title_canon_sha256":"a4a5f43679d3e94f6e06361665ca378d546a6fe1fd264a3934590f9378a3d406"},"schema_version":"1.0","source":{"id":"2605.14141","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14141","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14141v1","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14141","created_at":"2026-05-17T23:39:11Z"},{"alias_kind":"pith_short_12","alias_value":"MV3FB4SECH3O","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"MV3FB4SECH3OERER","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"MV3FB4SE","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:db418f933981ff32be1140e2e4ed5981897c7630d04f4eacb90c9f2a046308fc","target":"graph","created_at":"2026-05-17T23:39:11Z","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":"We prove that the empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime, and that statistically identifiable hints can be recovered and compiled from polynomially many samples."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That statistically identifiable solver hints exist for the target distributions and can be recovered from polynomially many samples to produce a fixed library of solvers whose fastest member generalizes in runtime."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"LLM agents synthesize distribution-aware solver code that achieves 0.971 mean normalized quality and runs 336x faster than top heuristics on 21 combinatorial optimization distributions."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"The empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime to new instances from the same distribution."}],"snapshot_sha256":"f36db2151c12d8105bdff8d72cf2e1e215409e81b7a5bd921e7ba5a0343a080d"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"982a0fab1c9f213162ee85cd2b8d0eb4df7f4973992311d9a18ad5c82d720aa3"},"paper":{"abstract_excerpt":"We study learning when the learned object is executable solver code rather than a predictor. In this setting, correctness is not enough: two solvers may both return valid solutions on the deployment distribution while differing substantially in runtime. Given samples from an unknown task distribution, the learner returns code evaluated on fresh instances by both solution quality and execution time. Our central abstraction is a \\emph{solver hint}: reusable structure inferred from samples and compiled into specialized solver code. We prove that the empirically fastest sample-consistent solver fr","authors_text":"Pierfrancesco Beneventano, Priyadarsi Mishra, Saharsh Koganti, Tomer Galanti","cross_cats":[],"headline":"The empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime to new instances from the same distribution.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T21:47:55Z","title":"Distribution-Aware Algorithm Design with LLM Agents"},"references":{"count":59,"internal_anchors":0,"resolved_work":59,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"D. L. Applegate, R. E. Bixby, V . Chvátal, and W. J. Cook.The Traveling Salesman Problem: A Computational Study. Princeton Series in Applied Mathematics. Princeton University Press, 2006","work_id":"da3eb0be-b01d-4827-9ac8-1747e399b9ec","year":2006},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"F. Avellaneda. A short description of the solver EvalMaxSAT. In F. Bacchus, J. Berg, M. Järvisalo, and R. Martins, editors,MaxSAT Evaluation 2020: Solver and Benchmark Descriptions, pages 8–9, 2020","work_id":"d1c88265-eefa-4067-8276-93409ebadf16","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"M. Balog, A. L. Gaunt, M. Brockschmidt, S. Nowozin, and D. Tarlow. Deepcoder: Learning to write programs. InInternational Conference on Learning Representations, 2017","work_id":"4c4dd0a9-5ebd-49be-8f9a-1b1ca63949bd","year":2017},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"K. Bestuzheva, M. Besançon, W.-K. Chen, A. Chmiela, T. Donkiewicz, J. van Doornmalen, L. Eifler, O. Gaul, G. Gamrath, A. Gleixner, et al. Enabling research through the SCIP optimization suite 8.0.ACM ","work_id":"7db8e0b5-2232-4718-bc8f-e7e5427ffaf3","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"A. Bogdanov and L. Trevisan. Average-case complexity.Foundations and Trends in Theoretical Computer Science, 2(1):1–106, 2006","work_id":"14263d28-0c99-4063-afd5-4a3b0edfb8c0","year":2006}],"snapshot_sha256":"2f77fb2e5f7a114b88cab5aed36dcd24b82c48243e13f7f108993c4939ce8a42"},"source":{"id":"2605.14141","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T04:59:58.226238Z","id":"d0c02064-3723-4809-a565-2fae46ef4866","model_set":{"reader":"grok-4.3"},"one_line_summary":"LLM agents synthesize distribution-aware solver code that achieves 0.971 mean normalized quality and runs 336x faster than top heuristics on 21 combinatorial optimization distributions.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"The empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime to new instances from the same distribution.","strongest_claim":"We prove that the empirically fastest sample-consistent solver from a fixed library generalizes in both correctness and runtime, and that statistically identifiable hints can be recovered and compiled from polynomially many samples.","weakest_assumption":"That statistically identifiable solver hints exist for the target distributions and can be recovered from polynomially many samples to produce a fixed library of solvers whose fastest member generalizes in runtime."}},"verdict_id":"d0c02064-3723-4809-a565-2fae46ef4866"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:ba7aef28efd9f0ad897809c23363b2360b20bba735df793bd96bd0a27d2c7f50","target":"record","created_at":"2026-05-17T23:39:11Z","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":"0c338b5ee72081852eae22d591daca02331ef2758aa81e032626ff43fc4b8bf7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T21:47:55Z","title_canon_sha256":"a4a5f43679d3e94f6e06361665ca378d546a6fe1fd264a3934590f9378a3d406"},"schema_version":"1.0","source":{"id":"2605.14141","kind":"arxiv","version":1}},"canonical_sha256":"657650f24411f6e244913994ac6ee68da9d899b2b40242601aafbb71ee298f33","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"657650f24411f6e244913994ac6ee68da9d899b2b40242601aafbb71ee298f33","first_computed_at":"2026-05-17T23:39:11.687677Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:11.687677Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jCxdTz46R6DeRwNYrTszg1RFLB4MtdJEOq/qh1mZiSyGeugbjrBQXSBV8HPUKjDw+t5mmbIYnd5n62rRMQuiDg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:11.688177Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14141","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ba7aef28efd9f0ad897809c23363b2360b20bba735df793bd96bd0a27d2c7f50","sha256:db418f933981ff32be1140e2e4ed5981897c7630d04f4eacb90c9f2a046308fc"],"state_sha256":"f1846baff78ad872a996a35ab77074fe8a34081dbcafe8f00104c77bb4034c54"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wG7PyxKFtyaU0yWoQyNm8Z2gEhYBBbZ4k4CKIckhSzIVss5bB83koLgOdDzHhKHYv96EUKsk4H/wLhZKiFxzDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-02T06:08:35.304489Z","bundle_sha256":"43ebd769c98bdc5c7af26c2a5a26ef50d6871b37d7024f444783c670667918fc"}}