{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4BVJROPAXSUCTRS5KZBKUQADBN","short_pith_number":"pith:4BVJROPA","canonical_record":{"source":{"id":"2605.12667","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T19:17:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"587504c2283391984dcacc0c61ac7c4a3ba7a095f7037440aca38a8efea03726","abstract_canon_sha256":"20cf49d7c089d5de6656ff33d2420fcd4c87b9714b87d760d8a9a206e4596a4b"},"schema_version":"1.0"},"canonical_sha256":"e06a98b9e0bca829c65d5642aa40030b57466e5bdfc12b0321aadb8be9a211cf","source":{"kind":"arxiv","id":"2605.12667","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12667","created_at":"2026-05-18T03:09:50Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12667v1","created_at":"2026-05-18T03:09:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12667","created_at":"2026-05-18T03:09:50Z"},{"alias_kind":"pith_short_12","alias_value":"4BVJROPAXSUC","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"4BVJROPAXSUCTRS5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"4BVJROPA","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4BVJROPAXSUCTRS5KZBKUQADBN","target":"record","payload":{"canonical_record":{"source":{"id":"2605.12667","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T19:17:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"587504c2283391984dcacc0c61ac7c4a3ba7a095f7037440aca38a8efea03726","abstract_canon_sha256":"20cf49d7c089d5de6656ff33d2420fcd4c87b9714b87d760d8a9a206e4596a4b"},"schema_version":"1.0"},"canonical_sha256":"e06a98b9e0bca829c65d5642aa40030b57466e5bdfc12b0321aadb8be9a211cf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:50.305122Z","signature_b64":"QgABlwWLNG1ucfjlj4Ux29YmUqTkTyGzBTmyzOnwmQyQZRHgTHfHp5B3ZQ4Ah44xPaWS1HxLl6V4fHOFHBCpBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e06a98b9e0bca829c65d5642aa40030b57466e5bdfc12b0321aadb8be9a211cf","last_reissued_at":"2026-05-18T03:09:50.304220Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:50.304220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.12667","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":"2026-05-18T03:09:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J1DaEYAtakcKpJ+a20Oa9aOoJAjKWBF2ozWs/DuznOf4jTXB49Etj+ZgA/8kvwipJUgJUZIWkwbYbZsVp/TuAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T15:54:52.744889Z"},"content_sha256":"08f98f199017d6fbd6a071058d33cbfb2c4d51f78a6e1dcb5a08568a1cffadc4","schema_version":"1.0","event_id":"sha256:08f98f199017d6fbd6a071058d33cbfb2c4d51f78a6e1dcb5a08568a1cffadc4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4BVJROPAXSUCTRS5KZBKUQADBN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ODRPO: Ordinal Decompositions of Discrete Rewards for Robust Policy Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Decomposing discrete rewards into ordinal binary indicators isolates evaluation noise and stabilizes policy updates in RLAIF without extra compute.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fei Wang, Inderjit Dhillon, Nirmal Patel","submitted_at":"2026-05-12T19:17:14Z","abstract_excerpt":"The alignment of Large Language Models (LLMs) utilizes Reinforcement Learning from AI Feedback (RLAIF) for non-verifiable domains such as long-form question answering and open-ended instruction following. These domains often rely on LLM based auto-raters to provide granular, multi-tier discrete rewards (e.g., 1-10 rubrics) that are inherently stochastic due to prompt sensitivity and sampling randomness. We empirically verify the stochasticity of auto-raters that can propagate and corrupt standard advantage estimators like GRPO and MaxRL, as a noisy reward samples can skew normalization statist"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"ODRPO achieves robust performance on Qwen2.5-7B and Qwen3-4B models, outperforming baselines with relative improvements of upto 14.8% on FACTS-grounding-v2 and 7.5% on Alpaca-Evals. Critically, these gains are achieved with negligible training-time overhead, as ODRPO requires no additional compute per step compared to standard estimators.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That decomposing discrete rewards into ordinal binary indicators structurally isolates evaluation noise and prevents outlier evaluations from corrupting the global update, as stated in the abstract description of the framework.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ODRPO decomposes discrete rewards into ordinal binary indicators to compute independent advantages and reduce noise corruption in RLAIF policy optimization.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Decomposing discrete rewards into ordinal binary indicators isolates evaluation noise and stabilizes policy updates in RLAIF without extra compute.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"98a7cb3a2c2621f4d7630879b5e9eb097a0ae8b35271b05abecc38f1c6e4bc6c"},"source":{"id":"2605.12667","kind":"arxiv","version":1},"verdict":{"id":"3589802d-eb98-4d60-a062-08a520eeac62","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:13:13.575935Z","strongest_claim":"ODRPO achieves robust performance on Qwen2.5-7B and Qwen3-4B models, outperforming baselines with relative improvements of upto 14.8% on FACTS-grounding-v2 and 7.5% on Alpaca-Evals. Critically, these gains are achieved with negligible training-time overhead, as ODRPO requires no additional compute per step compared to standard estimators.","one_line_summary":"ODRPO decomposes discrete rewards into ordinal binary indicators to compute independent advantages and reduce noise corruption in RLAIF policy optimization.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That decomposing discrete rewards into ordinal binary indicators structurally isolates evaluation noise and prevents outlier evaluations from corrupting the global update, as stated in the abstract description of the framework.","pith_extraction_headline":"Decomposing discrete rewards into ordinal binary indicators isolates evaluation noise and stabilizes policy updates in RLAIF without extra compute."},"references":{"count":41,"sample":[{"doi":"10.1038/s41586-025-09422-z","year":null,"title":"Nature645(8081), 633–638 (2025) https://doi.org/10.1038/s41586-025-09422-z","work_id":"9835b482-5032-4135-93dd-82a066677569","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles , author=. 2025 , eprint=","work_id":"7ea2960e-3de4-4048-be17-ddc3bb8327a6","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Hansen and Duo Peng and Yuhui Zhang and Alejandro Lozano and Min Woo Sun and Emma Lundberg and Serena Yeung-Levy , year=","work_id":"24cdde52-11e0-4409-8656-4a46c92a8552","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs , author=. 2025 , eprint=","work_id":"08971129-3986-43de-9574-8e903aff78da","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2026,"title":"Examining Reasoning LLMs-as-Judges in Non-Verifiable LLM Post-Training , author=. 2026 , eprint=","work_id":"01f55215-0262-4791-87f3-0a2c1cd0bf25","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":41,"snapshot_sha256":"4a64a2eec651d75f8ad382c46ecf342602c7241135526e5dbb9583aacf526677","internal_anchors":2},"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":"3589802d-eb98-4d60-a062-08a520eeac62"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HIM2MpHw0Qkx8utrDan/JHn+1OrWCmng5NZGUu7v/IJq95yhkOB59leGd7nc8saoKHIQ0jlZc2UTO9ezRXzvCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T15:54:52.745466Z"},"content_sha256":"b16b6743ae3257c739cfa5cb5a985454e97836611894e2159d0948b05306dd75","schema_version":"1.0","event_id":"sha256:b16b6743ae3257c739cfa5cb5a985454e97836611894e2159d0948b05306dd75"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4BVJROPAXSUCTRS5KZBKUQADBN/bundle.json","state_url":"https://pith.science/pith/4BVJROPAXSUCTRS5KZBKUQADBN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4BVJROPAXSUCTRS5KZBKUQADBN/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":"2026-06-01T15:54:52Z","links":{"resolver":"https://pith.science/pith/4BVJROPAXSUCTRS5KZBKUQADBN","bundle":"https://pith.science/pith/4BVJROPAXSUCTRS5KZBKUQADBN/bundle.json","state":"https://pith.science/pith/4BVJROPAXSUCTRS5KZBKUQADBN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4BVJROPAXSUCTRS5KZBKUQADBN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4BVJROPAXSUCTRS5KZBKUQADBN","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":"20cf49d7c089d5de6656ff33d2420fcd4c87b9714b87d760d8a9a206e4596a4b","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T19:17:14Z","title_canon_sha256":"587504c2283391984dcacc0c61ac7c4a3ba7a095f7037440aca38a8efea03726"},"schema_version":"1.0","source":{"id":"2605.12667","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12667","created_at":"2026-05-18T03:09:50Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12667v1","created_at":"2026-05-18T03:09:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12667","created_at":"2026-05-18T03:09:50Z"},{"alias_kind":"pith_short_12","alias_value":"4BVJROPAXSUC","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"4BVJROPAXSUCTRS5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"4BVJROPA","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:b16b6743ae3257c739cfa5cb5a985454e97836611894e2159d0948b05306dd75","target":"graph","created_at":"2026-05-18T03:09:50Z","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":"ODRPO achieves robust performance on Qwen2.5-7B and Qwen3-4B models, outperforming baselines with relative improvements of upto 14.8% on FACTS-grounding-v2 and 7.5% on Alpaca-Evals. Critically, these gains are achieved with negligible training-time overhead, as ODRPO requires no additional compute per step compared to standard estimators."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That decomposing discrete rewards into ordinal binary indicators structurally isolates evaluation noise and prevents outlier evaluations from corrupting the global update, as stated in the abstract description of the framework."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ODRPO decomposes discrete rewards into ordinal binary indicators to compute independent advantages and reduce noise corruption in RLAIF policy optimization."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Decomposing discrete rewards into ordinal binary indicators isolates evaluation noise and stabilizes policy updates in RLAIF without extra compute."}],"snapshot_sha256":"98a7cb3a2c2621f4d7630879b5e9eb097a0ae8b35271b05abecc38f1c6e4bc6c"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"The alignment of Large Language Models (LLMs) utilizes Reinforcement Learning from AI Feedback (RLAIF) for non-verifiable domains such as long-form question answering and open-ended instruction following. These domains often rely on LLM based auto-raters to provide granular, multi-tier discrete rewards (e.g., 1-10 rubrics) that are inherently stochastic due to prompt sensitivity and sampling randomness. We empirically verify the stochasticity of auto-raters that can propagate and corrupt standard advantage estimators like GRPO and MaxRL, as a noisy reward samples can skew normalization statist","authors_text":"Fei Wang, Inderjit Dhillon, Nirmal Patel","cross_cats":["cs.AI"],"headline":"Decomposing discrete rewards into ordinal binary indicators isolates evaluation noise and stabilizes policy updates in RLAIF without extra compute.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T19:17:14Z","title":"ODRPO: Ordinal Decompositions of Discrete Rewards for Robust Policy Optimization"},"references":{"count":41,"internal_anchors":2,"resolved_work":41,"sample":[{"cited_arxiv_id":"","doi":"10.1038/s41586-025-09422-z","is_internal_anchor":false,"ref_index":1,"title":"Nature645(8081), 633–638 (2025) https://doi.org/10.1038/s41586-025-09422-z","work_id":"9835b482-5032-4135-93dd-82a066677569","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles , author=. 2025 , eprint=","work_id":"7ea2960e-3de4-4048-be17-ddc3bb8327a6","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Hansen and Duo Peng and Yuhui Zhang and Alejandro Lozano and Min Woo Sun and Emma Lundberg and Serena Yeung-Levy , year=","work_id":"24cdde52-11e0-4409-8656-4a46c92a8552","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"ViCrit: A Verifiable Reinforcement Learning Proxy Task for Visual Perception in VLMs , author=. 2025 , eprint=","work_id":"08971129-3986-43de-9574-8e903aff78da","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Examining Reasoning LLMs-as-Judges in Non-Verifiable LLM Post-Training , author=. 2026 , eprint=","work_id":"01f55215-0262-4791-87f3-0a2c1cd0bf25","year":2026}],"snapshot_sha256":"4a64a2eec651d75f8ad382c46ecf342602c7241135526e5dbb9583aacf526677"},"source":{"id":"2605.12667","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T21:13:13.575935Z","id":"3589802d-eb98-4d60-a062-08a520eeac62","model_set":{"reader":"grok-4.3"},"one_line_summary":"ODRPO decomposes discrete rewards into ordinal binary indicators to compute independent advantages and reduce noise corruption in RLAIF policy optimization.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Decomposing discrete rewards into ordinal binary indicators isolates evaluation noise and stabilizes policy updates in RLAIF without extra compute.","strongest_claim":"ODRPO achieves robust performance on Qwen2.5-7B and Qwen3-4B models, outperforming baselines with relative improvements of upto 14.8% on FACTS-grounding-v2 and 7.5% on Alpaca-Evals. Critically, these gains are achieved with negligible training-time overhead, as ODRPO requires no additional compute per step compared to standard estimators.","weakest_assumption":"That decomposing discrete rewards into ordinal binary indicators structurally isolates evaluation noise and prevents outlier evaluations from corrupting the global update, as stated in the abstract description of the framework."}},"verdict_id":"3589802d-eb98-4d60-a062-08a520eeac62"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:08f98f199017d6fbd6a071058d33cbfb2c4d51f78a6e1dcb5a08568a1cffadc4","target":"record","created_at":"2026-05-18T03:09:50Z","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":"20cf49d7c089d5de6656ff33d2420fcd4c87b9714b87d760d8a9a206e4596a4b","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T19:17:14Z","title_canon_sha256":"587504c2283391984dcacc0c61ac7c4a3ba7a095f7037440aca38a8efea03726"},"schema_version":"1.0","source":{"id":"2605.12667","kind":"arxiv","version":1}},"canonical_sha256":"e06a98b9e0bca829c65d5642aa40030b57466e5bdfc12b0321aadb8be9a211cf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e06a98b9e0bca829c65d5642aa40030b57466e5bdfc12b0321aadb8be9a211cf","first_computed_at":"2026-05-18T03:09:50.304220Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:50.304220Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QgABlwWLNG1ucfjlj4Ux29YmUqTkTyGzBTmyzOnwmQyQZRHgTHfHp5B3ZQ4Ah44xPaWS1HxLl6V4fHOFHBCpBg==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:50.305122Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12667","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:08f98f199017d6fbd6a071058d33cbfb2c4d51f78a6e1dcb5a08568a1cffadc4","sha256:b16b6743ae3257c739cfa5cb5a985454e97836611894e2159d0948b05306dd75"],"state_sha256":"618fd78982e4764d51fd42a0bbf09799e350b119e55714795b3157900a1b9db6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dSd9aflxx7Ilt7fiyQ/sC723ps67qCw4As4k1KqVTgx1DyqDJ22oSA+pF5MXHydA9ClAPSgu++ibKmNlv+82CQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T15:54:52.747817Z","bundle_sha256":"aae28033cd7ae719aa2ce695be5540d59f805e5c267cc3933910c6ccaaea6597"}}