{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4YZWJ3NDD4KW5DD7IFVMSXB6CA","short_pith_number":"pith:4YZWJ3ND","canonical_record":{"source":{"id":"2605.04902","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DB","submitted_at":"2026-05-06T13:31:37Z","cross_cats_sorted":[],"title_canon_sha256":"d103d8a16c2c309530115cba0b4eff95c7a024902b9e44ace77449bf2f9b351f","abstract_canon_sha256":"a76b2863222eca801eac7edc2e4f4ce375b7980f682cc78a8d30b2ec6eeb0231"},"schema_version":"1.0"},"canonical_sha256":"e63364eda31f156e8c7f416ac95c3e103509d5b718673580f5d00f967148599d","source":{"kind":"arxiv","id":"2605.04902","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.04902","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"arxiv_version","alias_value":"2605.04902v3","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.04902","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"pith_short_12","alias_value":"4YZWJ3NDD4KW","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"pith_short_16","alias_value":"4YZWJ3NDD4KW5DD7","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"pith_short_8","alias_value":"4YZWJ3ND","created_at":"2026-06-02T01:04:16Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4YZWJ3NDD4KW5DD7IFVMSXB6CA","target":"record","payload":{"canonical_record":{"source":{"id":"2605.04902","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DB","submitted_at":"2026-05-06T13:31:37Z","cross_cats_sorted":[],"title_canon_sha256":"d103d8a16c2c309530115cba0b4eff95c7a024902b9e44ace77449bf2f9b351f","abstract_canon_sha256":"a76b2863222eca801eac7edc2e4f4ce375b7980f682cc78a8d30b2ec6eeb0231"},"schema_version":"1.0"},"canonical_sha256":"e63364eda31f156e8c7f416ac95c3e103509d5b718673580f5d00f967148599d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:04:16.502727Z","signature_b64":"dDsKBOdSBs2Qg/eWO6tcppPJ/cunW05ykf7qd6Xe0XXaNnN7S58PtJ8ySt2ZLVHyG3n6ShJBtbtvzph902S5Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e63364eda31f156e8c7f416ac95c3e103509d5b718673580f5d00f967148599d","last_reissued_at":"2026-06-02T01:04:16.502178Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:04:16.502178Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.04902","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:04:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LtTFGlgt1KKRQ93XT6PEuKnsnc/W5a8Twtc05Wl6AyYL22H62rh2NIDzD+WcWETrECCPSoZppNEdXdM/EHWiCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T10:23:03.083043Z"},"content_sha256":"a5ce7304272d7ab50ce6ca6b01690e4d94e872c560c55abff5cceb0da22803f1","schema_version":"1.0","event_id":"sha256:a5ce7304272d7ab50ce6ca6b01690e4d94e872c560c55abff5cceb0da22803f1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4YZWJ3NDD4KW5DD7IFVMSXB6CA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AegisTS: A Hierarchical Agent System with Reinforcement Learning for Multivariate Time Series Data Cleaning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A hierarchical reinforcement learning agent system can jointly optimize the processing order and method selection to clean multiple quality issues in multivariate time series data without needing ground truth.","cross_cats":[],"primary_cat":"cs.DB","authors_text":"Lu Chen, Mourad Khayati, Tianyi Li, Yuanyuan Yao, Yuhan Shi","submitted_at":"2026-05-06T13:31:37Z","abstract_excerpt":"Multivariate time series (MTS) are frequently affected by co-occurring quality issues, such as missing values, outliers, and constraint violations, which significantly undermine downstream analytics. Existing cleaning approaches fix only a limited set of such issues, making them ill-suited for scenarios where multiple quality problems arise simultaneously. Furthermore, these methods commonly depend on the availability of ground truth data or domain-specific rules, both of which are rarely accessible in real-world applications.\n  In this paper, we introduce AegisTS, an agent system with reinfor"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our experimental results show that AegisTS consistently outperforms existing methods, achieving up to 96% improvement in data cleaning quality and 27% improvement in downstream performance.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the proposed dual-stage reward mechanism, which couples upstream cleaning quality with downstream task performance, can reliably guide the hierarchical agents to optimal pipelines in the absence of ground truth data.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"AegisTS uses a two-level RL agent architecture with a dual-stage reward to jointly optimize cleaning order and method selection for multivariate time series, delivering up to 96% better cleaning quality and 27% better downstream performance without ground truth.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A hierarchical reinforcement learning agent system can jointly optimize the processing order and method selection to clean multiple quality issues in multivariate time series data without needing ground truth.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4f5f82bb40717c7614ffcfde57febc050132e3c21505ced589cd2b9758e9dde1"},"source":{"id":"2605.04902","kind":"arxiv","version":3},"verdict":{"id":"d0cef379-a393-4ef4-9089-ba7c95790684","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T16:21:01.387848Z","strongest_claim":"Our experimental results show that AegisTS consistently outperforms existing methods, achieving up to 96% improvement in data cleaning quality and 27% improvement in downstream performance.","one_line_summary":"AegisTS uses a two-level RL agent architecture with a dual-stage reward to jointly optimize cleaning order and method selection for multivariate time series, delivering up to 96% better cleaning quality and 27% better downstream performance without ground truth.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the proposed dual-stage reward mechanism, which couples upstream cleaning quality with downstream task performance, can reliably guide the hierarchical agents to optimal pipelines in the absence of ground truth data.","pith_extraction_headline":"A hierarchical reinforcement learning agent system can jointly optimize the processing order and method selection to clean multiple quality issues in multivariate time series data without needing ground truth."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.04902/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T10:41:00.159384Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:23.702943Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T14:03:43.831112Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"4424964b1041675537bdecce4eecd371dd85fafd526b7a27c7179d95a6bf0089"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"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":"d0cef379-a393-4ef4-9089-ba7c95790684"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:04:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nLX2MCowf4wSX0sd2OYcKNgpYcVlqrWVeW6NRMUHYBMU3NKHw5cASNlLaAvL5dPZN0sINekJkO+7vqZ4UNaKCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T10:23:03.083744Z"},"content_sha256":"bcca94f8e61b243098a0a22ebc4f12b2dd750b48afc541435ca797bac4739ff6","schema_version":"1.0","event_id":"sha256:bcca94f8e61b243098a0a22ebc4f12b2dd750b48afc541435ca797bac4739ff6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4YZWJ3NDD4KW5DD7IFVMSXB6CA/bundle.json","state_url":"https://pith.science/pith/4YZWJ3NDD4KW5DD7IFVMSXB6CA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4YZWJ3NDD4KW5DD7IFVMSXB6CA/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":"2026-06-07T10:23:03Z","links":{"resolver":"https://pith.science/pith/4YZWJ3NDD4KW5DD7IFVMSXB6CA","bundle":"https://pith.science/pith/4YZWJ3NDD4KW5DD7IFVMSXB6CA/bundle.json","state":"https://pith.science/pith/4YZWJ3NDD4KW5DD7IFVMSXB6CA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4YZWJ3NDD4KW5DD7IFVMSXB6CA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4YZWJ3NDD4KW5DD7IFVMSXB6CA","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":"a76b2863222eca801eac7edc2e4f4ce375b7980f682cc78a8d30b2ec6eeb0231","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DB","submitted_at":"2026-05-06T13:31:37Z","title_canon_sha256":"d103d8a16c2c309530115cba0b4eff95c7a024902b9e44ace77449bf2f9b351f"},"schema_version":"1.0","source":{"id":"2605.04902","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.04902","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"arxiv_version","alias_value":"2605.04902v3","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.04902","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"pith_short_12","alias_value":"4YZWJ3NDD4KW","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"pith_short_16","alias_value":"4YZWJ3NDD4KW5DD7","created_at":"2026-06-02T01:04:16Z"},{"alias_kind":"pith_short_8","alias_value":"4YZWJ3ND","created_at":"2026-06-02T01:04:16Z"}],"graph_snapshots":[{"event_id":"sha256:bcca94f8e61b243098a0a22ebc4f12b2dd750b48afc541435ca797bac4739ff6","target":"graph","created_at":"2026-06-02T01:04:16Z","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 experimental results show that AegisTS consistently outperforms existing methods, achieving up to 96% improvement in data cleaning quality and 27% improvement in downstream performance."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the proposed dual-stage reward mechanism, which couples upstream cleaning quality with downstream task performance, can reliably guide the hierarchical agents to optimal pipelines in the absence of ground truth data."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"AegisTS uses a two-level RL agent architecture with a dual-stage reward to jointly optimize cleaning order and method selection for multivariate time series, delivering up to 96% better cleaning quality and 27% better downstream performance without ground truth."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A hierarchical reinforcement learning agent system can jointly optimize the processing order and method selection to clean multiple quality issues in multivariate time series data without needing ground truth."}],"snapshot_sha256":"4f5f82bb40717c7614ffcfde57febc050132e3c21505ced589cd2b9758e9dde1"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T10:41:00.159384Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T22:01:23.702943Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T14:03:43.831112Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.04902/integrity.json","findings":[],"snapshot_sha256":"4424964b1041675537bdecce4eecd371dd85fafd526b7a27c7179d95a6bf0089","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multivariate time series (MTS) are frequently affected by co-occurring quality issues, such as missing values, outliers, and constraint violations, which significantly undermine downstream analytics. Existing cleaning approaches fix only a limited set of such issues, making them ill-suited for scenarios where multiple quality problems arise simultaneously. Furthermore, these methods commonly depend on the availability of ground truth data or domain-specific rules, both of which are rarely accessible in real-world applications.\n  In this paper, we introduce AegisTS, an agent system with reinfor","authors_text":"Lu Chen, Mourad Khayati, Tianyi Li, Yuanyuan Yao, Yuhan Shi","cross_cats":[],"headline":"A hierarchical reinforcement learning agent system can jointly optimize the processing order and method selection to clean multiple quality issues in multivariate time series data without needing ground truth.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DB","submitted_at":"2026-05-06T13:31:37Z","title":"AegisTS: A Hierarchical Agent System with Reinforcement Learning for Multivariate Time Series Data Cleaning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.04902","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-08T16:21:01.387848Z","id":"d0cef379-a393-4ef4-9089-ba7c95790684","model_set":{"reader":"grok-4.3"},"one_line_summary":"AegisTS uses a two-level RL agent architecture with a dual-stage reward to jointly optimize cleaning order and method selection for multivariate time series, delivering up to 96% better cleaning quality and 27% better downstream performance without ground truth.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A hierarchical reinforcement learning agent system can jointly optimize the processing order and method selection to clean multiple quality issues in multivariate time series data without needing ground truth.","strongest_claim":"Our experimental results show that AegisTS consistently outperforms existing methods, achieving up to 96% improvement in data cleaning quality and 27% improvement in downstream performance.","weakest_assumption":"That the proposed dual-stage reward mechanism, which couples upstream cleaning quality with downstream task performance, can reliably guide the hierarchical agents to optimal pipelines in the absence of ground truth data."}},"verdict_id":"d0cef379-a393-4ef4-9089-ba7c95790684"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:a5ce7304272d7ab50ce6ca6b01690e4d94e872c560c55abff5cceb0da22803f1","target":"record","created_at":"2026-06-02T01:04:16Z","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":"a76b2863222eca801eac7edc2e4f4ce375b7980f682cc78a8d30b2ec6eeb0231","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DB","submitted_at":"2026-05-06T13:31:37Z","title_canon_sha256":"d103d8a16c2c309530115cba0b4eff95c7a024902b9e44ace77449bf2f9b351f"},"schema_version":"1.0","source":{"id":"2605.04902","kind":"arxiv","version":3}},"canonical_sha256":"e63364eda31f156e8c7f416ac95c3e103509d5b718673580f5d00f967148599d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e63364eda31f156e8c7f416ac95c3e103509d5b718673580f5d00f967148599d","first_computed_at":"2026-06-02T01:04:16.502178Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:04:16.502178Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"dDsKBOdSBs2Qg/eWO6tcppPJ/cunW05ykf7qd6Xe0XXaNnN7S58PtJ8ySt2ZLVHyG3n6ShJBtbtvzph902S5Aw==","signature_status":"signed_v1","signed_at":"2026-06-02T01:04:16.502727Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.04902","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a5ce7304272d7ab50ce6ca6b01690e4d94e872c560c55abff5cceb0da22803f1","sha256:bcca94f8e61b243098a0a22ebc4f12b2dd750b48afc541435ca797bac4739ff6"],"state_sha256":"e2006cd38ce0ce9b6150b3a2758b64af49f762308529236b82d080baf461c299"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6wLxQ/gdCw44vCCziHMprMzCao23Qvw5K7E/OoHb/MdzzRDxZSkkiw/izXdVgpmJvMEmH6MsMqm42Jtdk1baBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-07T10:23:03.086881Z","bundle_sha256":"91bfd76bc1baabd18b0cf7f8aa325af6f6ec39b39a9a83f723f93d05902faeef"}}