{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:HXICGPKPZ5QHLYIDZNDQKUNONG","short_pith_number":"pith:HXICGPKP","canonical_record":{"source":{"id":"2602.05408","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-02-05T07:45:42Z","cross_cats_sorted":[],"title_canon_sha256":"c251f592e2868d8fa68c79b7c523b3c4cc30ee8a0f3f57a4623a62fd550686ed","abstract_canon_sha256":"a7fd24e7f97b21ef428f7a9a29a546f4371968ddec51bf6e373b714004243f5b"},"schema_version":"1.0"},"canonical_sha256":"3dd0233d4fcf6075e103cb470551ae69abff64536eee7c0e851f363c7bf385a1","source":{"kind":"arxiv","id":"2602.05408","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.05408","created_at":"2026-05-18T02:44:31Z"},{"alias_kind":"arxiv_version","alias_value":"2602.05408v2","created_at":"2026-05-18T02:44:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.05408","created_at":"2026-05-18T02:44:31Z"},{"alias_kind":"pith_short_12","alias_value":"HXICGPKPZ5QH","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"HXICGPKPZ5QHLYID","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"HXICGPKP","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:HXICGPKPZ5QHLYIDZNDQKUNONG","target":"record","payload":{"canonical_record":{"source":{"id":"2602.05408","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-02-05T07:45:42Z","cross_cats_sorted":[],"title_canon_sha256":"c251f592e2868d8fa68c79b7c523b3c4cc30ee8a0f3f57a4623a62fd550686ed","abstract_canon_sha256":"a7fd24e7f97b21ef428f7a9a29a546f4371968ddec51bf6e373b714004243f5b"},"schema_version":"1.0"},"canonical_sha256":"3dd0233d4fcf6075e103cb470551ae69abff64536eee7c0e851f363c7bf385a1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:31.592005Z","signature_b64":"r3ogaO4ZL3HYu3kqNi/UBfkNeyd1QoATf09vtWdCRFR3+RbQ+HL9j8rN7801Y6FgaEx7d39WPXLKrfT3M+ZiBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3dd0233d4fcf6075e103cb470551ae69abff64536eee7c0e851f363c7bf385a1","last_reissued_at":"2026-05-18T02:44:31.591454Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:31.591454Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.05408","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":"2026-05-18T02:44:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B4622VHi65bo/5NgtrjwfbdDYUdFBr45ncb+iiQHohsfZBRz98pHqU3UyaA51LspTjipPyGztD5iKZ1qsTfGAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T09:05:19.640625Z"},"content_sha256":"9afcb6039b665eae5b6a5fc0405831c9c8b386fea40a565d3d136f866f2a87ab","schema_version":"1.0","event_id":"sha256:9afcb6039b665eae5b6a5fc0405831c9c8b386fea40a565d3d136f866f2a87ab"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:HXICGPKPZ5QHLYIDZNDQKUNONG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rich-Media Re-Ranker: A User Satisfaction-Driven LLM Re-ranking Framework for Rich-Media Search","license":"http://creativecommons.org/licenses/by/4.0/","headline":"The Rich-Media Re-Ranker decomposes session queries into sub-queries and uses an LLM evaluator informed by VLM visual signals to score results across relevance, quality, novelty, information gain, and presentation for higher user search满意度.","cross_cats":[],"primary_cat":"cs.IR","authors_text":"Feicheng Li, Jianxin Li, Ligang Zhou, Qingyun Sun, Ying Nie, Zeyang Tang, Zhiming Peng, Zihao Guo","submitted_at":"2026-02-05T07:45:42Z","abstract_excerpt":"Re-ranking plays a crucial role in modern information search systems by refining the ranking of initial search results to better satisfy user information needs. However, existing methods show two notable limitations in improving user search satisfaction: inadequate modeling of multifaceted user intents and neglect of rich side information such as visual perception signals. To address these challenges, we propose the Rich-Media Re-Ranker framework, which aims to enhance user search satisfaction through multi-dimensional and fine-grained modeling. Our approach begins with a Query Planner that an"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Extensive experiments demonstrate that our method significantly outperforms state-of-the-art baselines. Notably, the proposed framework has been deployed in a large-scale industrial search system, yielding substantial improvements in online user engagement rates and satisfaction metrics.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the VLM-based evaluator and LLM-based re-ranker, after multi-task reinforcement learning, produce reliable holistic scores on relevance, quality, novelty, information gain, and visual presentation without systematic bias or overfitting to the training distribution.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A re-ranking system for rich-media search that plans query intents from sessions, adds visual signals from VLMs, and uses an LLM to score results on multiple facets before multi-task RL adaptation, with reported gains in engagement after industrial deployment.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"The Rich-Media Re-Ranker decomposes session queries into sub-queries and uses an LLM evaluator informed by VLM visual signals to score results across relevance, quality, novelty, information gain, and presentation for higher user search满意度.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"398f2365ed2e62faa550f84d90371736a65b4a0ba6a249be0ffa7b7c3c2ac9c5"},"source":{"id":"2602.05408","kind":"arxiv","version":2},"verdict":{"id":"862a82b5-598d-4025-9b31-d6f1418a4645","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T07:27:29.587057Z","strongest_claim":"Extensive experiments demonstrate that our method significantly outperforms state-of-the-art baselines. Notably, the proposed framework has been deployed in a large-scale industrial search system, yielding substantial improvements in online user engagement rates and satisfaction metrics.","one_line_summary":"A re-ranking system for rich-media search that plans query intents from sessions, adds visual signals from VLMs, and uses an LLM to score results on multiple facets before multi-task RL adaptation, with reported gains in engagement after industrial deployment.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the VLM-based evaluator and LLM-based re-ranker, after multi-task reinforcement learning, produce reliable holistic scores on relevance, quality, novelty, information gain, and visual presentation without systematic bias or overfitting to the training distribution.","pith_extraction_headline":"The Rich-Media Re-Ranker decomposes session queries into sub-queries and uses an LLM evaluator informed by VLM visual signals to score results across relevance, quality, novelty, information gain, and presentation for higher user search满意度."},"references":{"count":50,"sample":[{"doi":"","year":null,"title":"Abdelrahman Abdallah, Jamshid Mozafari, Bhawna Piryani, and Adam Jatowt","work_id":"de94de6d-c4ef-49d7-be0c-81f002315741","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Dear: Dual-stage document reranking with reasoning agents via llm distillation","work_id":"93fae652-5f30-4502-8c99-d76286e8d25a","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Qingyao Ai, Keping Bi, Jiafeng Guo, and W Bruce Croft. 2018. Learning a deep listwise context model for ranking refinement. InThe 41st international ACM SIGIR conference on research & development in i","work_id":"bb87a37b-2069-4ebb-82ee-9f00b5d1689a","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Seq2slate: Re-ranking and slate optimization with rnns","work_id":"9868df2b-b11e-4c02-9e68-d8d1d535f9a2","ref_index":4,"cited_arxiv_id":"1810.02019","is_internal_anchor":true},{"doi":"","year":null,"title":"Heng-Tze Cheng, Levent Koc, Jeremiah Harmsen, Tal Shaked, Tushar Chandra, Hrishi Aradhye, Glen Anderson, Greg Corrado, Wei Chai, Mustafa Ispir, et al","work_id":"2276a040-1de3-4af7-9cca-f0a38c0eb0c7","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":50,"snapshot_sha256":"7bd9998d1e8940abc5e601abb81ee313600041c8536b6c5b5b65e9aa2e127e80","internal_anchors":8},"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":"862a82b5-598d-4025-9b31-d6f1418a4645"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4QJxOWufXmO9W53w8QjvcZdwuB4F5opiq6KqYw+ud7r8j2/RkQnI8WWgpndffxckQJzY1s9q2pwe36rvcS9zAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T09:05:19.641475Z"},"content_sha256":"c3aad53745b99cfab89acb51efa3c5d526e40ec4e605b9fcd6fe45dd55b4ed64","schema_version":"1.0","event_id":"sha256:c3aad53745b99cfab89acb51efa3c5d526e40ec4e605b9fcd6fe45dd55b4ed64"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HXICGPKPZ5QHLYIDZNDQKUNONG/bundle.json","state_url":"https://pith.science/pith/HXICGPKPZ5QHLYIDZNDQKUNONG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HXICGPKPZ5QHLYIDZNDQKUNONG/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":"2026-06-07T09:05:19Z","links":{"resolver":"https://pith.science/pith/HXICGPKPZ5QHLYIDZNDQKUNONG","bundle":"https://pith.science/pith/HXICGPKPZ5QHLYIDZNDQKUNONG/bundle.json","state":"https://pith.science/pith/HXICGPKPZ5QHLYIDZNDQKUNONG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HXICGPKPZ5QHLYIDZNDQKUNONG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:HXICGPKPZ5QHLYIDZNDQKUNONG","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":"a7fd24e7f97b21ef428f7a9a29a546f4371968ddec51bf6e373b714004243f5b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-02-05T07:45:42Z","title_canon_sha256":"c251f592e2868d8fa68c79b7c523b3c4cc30ee8a0f3f57a4623a62fd550686ed"},"schema_version":"1.0","source":{"id":"2602.05408","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.05408","created_at":"2026-05-18T02:44:31Z"},{"alias_kind":"arxiv_version","alias_value":"2602.05408v2","created_at":"2026-05-18T02:44:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.05408","created_at":"2026-05-18T02:44:31Z"},{"alias_kind":"pith_short_12","alias_value":"HXICGPKPZ5QH","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"HXICGPKPZ5QHLYID","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"HXICGPKP","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:c3aad53745b99cfab89acb51efa3c5d526e40ec4e605b9fcd6fe45dd55b4ed64","target":"graph","created_at":"2026-05-18T02:44:31Z","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":"Extensive experiments demonstrate that our method significantly outperforms state-of-the-art baselines. Notably, the proposed framework has been deployed in a large-scale industrial search system, yielding substantial improvements in online user engagement rates and satisfaction metrics."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the VLM-based evaluator and LLM-based re-ranker, after multi-task reinforcement learning, produce reliable holistic scores on relevance, quality, novelty, information gain, and visual presentation without systematic bias or overfitting to the training distribution."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A re-ranking system for rich-media search that plans query intents from sessions, adds visual signals from VLMs, and uses an LLM to score results on multiple facets before multi-task RL adaptation, with reported gains in engagement after industrial deployment."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"The Rich-Media Re-Ranker decomposes session queries into sub-queries and uses an LLM evaluator informed by VLM visual signals to score results across relevance, quality, novelty, information gain, and presentation for higher user search满意度."}],"snapshot_sha256":"398f2365ed2e62faa550f84d90371736a65b4a0ba6a249be0ffa7b7c3c2ac9c5"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Re-ranking plays a crucial role in modern information search systems by refining the ranking of initial search results to better satisfy user information needs. However, existing methods show two notable limitations in improving user search satisfaction: inadequate modeling of multifaceted user intents and neglect of rich side information such as visual perception signals. To address these challenges, we propose the Rich-Media Re-Ranker framework, which aims to enhance user search satisfaction through multi-dimensional and fine-grained modeling. Our approach begins with a Query Planner that an","authors_text":"Feicheng Li, Jianxin Li, Ligang Zhou, Qingyun Sun, Ying Nie, Zeyang Tang, Zhiming Peng, Zihao Guo","cross_cats":[],"headline":"The Rich-Media Re-Ranker decomposes session queries into sub-queries and uses an LLM evaluator informed by VLM visual signals to score results across relevance, quality, novelty, information gain, and presentation for higher user search满意度.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-02-05T07:45:42Z","title":"Rich-Media Re-Ranker: A User Satisfaction-Driven LLM Re-ranking Framework for Rich-Media Search"},"references":{"count":50,"internal_anchors":8,"resolved_work":50,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Abdelrahman Abdallah, Jamshid Mozafari, Bhawna Piryani, and Adam Jatowt","work_id":"de94de6d-c4ef-49d7-be0c-81f002315741","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Dear: Dual-stage document reranking with reasoning agents via llm distillation","work_id":"93fae652-5f30-4502-8c99-d76286e8d25a","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Qingyao Ai, Keping Bi, Jiafeng Guo, and W Bruce Croft. 2018. Learning a deep listwise context model for ranking refinement. InThe 41st international ACM SIGIR conference on research & development in i","work_id":"bb87a37b-2069-4ebb-82ee-9f00b5d1689a","year":2018},{"cited_arxiv_id":"1810.02019","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Seq2slate: Re-ranking and slate optimization with rnns","work_id":"9868df2b-b11e-4c02-9e68-d8d1d535f9a2","year":2018},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Heng-Tze Cheng, Levent Koc, Jeremiah Harmsen, Tal Shaked, Tushar Chandra, Hrishi Aradhye, Glen Anderson, Greg Corrado, Wei Chai, Mustafa Ispir, et al","work_id":"2276a040-1de3-4af7-9cca-f0a38c0eb0c7","year":null}],"snapshot_sha256":"7bd9998d1e8940abc5e601abb81ee313600041c8536b6c5b5b65e9aa2e127e80"},"source":{"id":"2602.05408","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-16T07:27:29.587057Z","id":"862a82b5-598d-4025-9b31-d6f1418a4645","model_set":{"reader":"grok-4.3"},"one_line_summary":"A re-ranking system for rich-media search that plans query intents from sessions, adds visual signals from VLMs, and uses an LLM to score results on multiple facets before multi-task RL adaptation, with reported gains in engagement after industrial deployment.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"The Rich-Media Re-Ranker decomposes session queries into sub-queries and uses an LLM evaluator informed by VLM visual signals to score results across relevance, quality, novelty, information gain, and presentation for higher user search满意度.","strongest_claim":"Extensive experiments demonstrate that our method significantly outperforms state-of-the-art baselines. Notably, the proposed framework has been deployed in a large-scale industrial search system, yielding substantial improvements in online user engagement rates and satisfaction metrics.","weakest_assumption":"That the VLM-based evaluator and LLM-based re-ranker, after multi-task reinforcement learning, produce reliable holistic scores on relevance, quality, novelty, information gain, and visual presentation without systematic bias or overfitting to the training distribution."}},"verdict_id":"862a82b5-598d-4025-9b31-d6f1418a4645"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9afcb6039b665eae5b6a5fc0405831c9c8b386fea40a565d3d136f866f2a87ab","target":"record","created_at":"2026-05-18T02:44:31Z","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":"a7fd24e7f97b21ef428f7a9a29a546f4371968ddec51bf6e373b714004243f5b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-02-05T07:45:42Z","title_canon_sha256":"c251f592e2868d8fa68c79b7c523b3c4cc30ee8a0f3f57a4623a62fd550686ed"},"schema_version":"1.0","source":{"id":"2602.05408","kind":"arxiv","version":2}},"canonical_sha256":"3dd0233d4fcf6075e103cb470551ae69abff64536eee7c0e851f363c7bf385a1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3dd0233d4fcf6075e103cb470551ae69abff64536eee7c0e851f363c7bf385a1","first_computed_at":"2026-05-18T02:44:31.591454Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:31.591454Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"r3ogaO4ZL3HYu3kqNi/UBfkNeyd1QoATf09vtWdCRFR3+RbQ+HL9j8rN7801Y6FgaEx7d39WPXLKrfT3M+ZiBg==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:31.592005Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.05408","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9afcb6039b665eae5b6a5fc0405831c9c8b386fea40a565d3d136f866f2a87ab","sha256:c3aad53745b99cfab89acb51efa3c5d526e40ec4e605b9fcd6fe45dd55b4ed64"],"state_sha256":"4269564a585f189ec69fffd340f641ab650e930856cc410872269910c62dfa0a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cMJFF/oNo0SuBRbBgrt7N2D+KlWrYi2MP82TS1bwUZsl6HBFOHqRNcu88tgeSQkJdzSPHZv+ePH8jmSS5svnDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-07T09:05:19.645173Z","bundle_sha256":"c812ebe3addaded064cfa8b2e0d486db09cb0ae402b2db4efaeef45d00d1bd89"}}