{"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"}