{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:7HTTNDGSPXUCM6OI6CFJBBJADB","short_pith_number":"pith:7HTTNDGS","schema_version":"1.0","canonical_sha256":"f9e7368cd27de82679c8f08a908520184fada131c96bf8cf37d02b8f396ba9dd","source":{"kind":"arxiv","id":"2410.10469","version":1},"attestation_state":"computed","paper":{"title":"Moirai-MoE: Empowering Time Series Foundation Models with Sparse Mixture of Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Caiming Xiong, Chenghao Liu, Doyen Sahoo, Gerald Woo, Juncheng Liu, Roger Zimmermann, Silvio Savarese, Taha Aksu, Xu Liu, Yuxuan Liang","submitted_at":"2024-10-14T13:01:11Z","abstract_excerpt":"Time series foundation models have demonstrated impressive performance as zero-shot forecasters. However, achieving effectively unified training on time series remains an open challenge. Existing approaches introduce some level of model specialization to account for the highly heterogeneous nature of time series data. For instance, Moirai pursues unified training by employing multiple input/output projection layers, each tailored to handle time series at a specific frequency. Similarly, TimesFM maintains a frequency embedding dictionary for this purpose. We identify two major drawbacks to this"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2410.10469","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-14T13:01:11Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"03385209caa8a64a544caf1d979d09ecb66682385d3ff0bd3da571eb79fec31d","abstract_canon_sha256":"053b0e7687b39a6dae72cf306cdb8ea37c58612e39e2eb7aa945e1ac4961edef"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:20:16.852871Z","signature_b64":"sJiVbSZtbSQqYTtH0h81lOnTtPmaNHWdQEZrcL1dpITCBGvYZVokVusVlGyhFJG9sa2ME3JIcaEQI5GTKkB8DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f9e7368cd27de82679c8f08a908520184fada131c96bf8cf37d02b8f396ba9dd","last_reissued_at":"2026-07-05T09:20:16.852430Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:20:16.852430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Moirai-MoE: Empowering Time Series Foundation Models with Sparse Mixture of Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Caiming Xiong, Chenghao Liu, Doyen Sahoo, Gerald Woo, Juncheng Liu, Roger Zimmermann, Silvio Savarese, Taha Aksu, Xu Liu, Yuxuan Liang","submitted_at":"2024-10-14T13:01:11Z","abstract_excerpt":"Time series foundation models have demonstrated impressive performance as zero-shot forecasters. However, achieving effectively unified training on time series remains an open challenge. Existing approaches introduce some level of model specialization to account for the highly heterogeneous nature of time series data. For instance, Moirai pursues unified training by employing multiple input/output projection layers, each tailored to handle time series at a specific frequency. Similarly, TimesFM maintains a frequency embedding dictionary for this purpose. We identify two major drawbacks to this"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.10469","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2410.10469/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2410.10469","created_at":"2026-07-05T09:20:16.852491+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.10469v1","created_at":"2026-07-05T09:20:16.852491+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.10469","created_at":"2026-07-05T09:20:16.852491+00:00"},{"alias_kind":"pith_short_12","alias_value":"7HTTNDGSPXUC","created_at":"2026-07-05T09:20:16.852491+00:00"},{"alias_kind":"pith_short_16","alias_value":"7HTTNDGSPXUCM6OI","created_at":"2026-07-05T09:20:16.852491+00:00"},{"alias_kind":"pith_short_8","alias_value":"7HTTNDGS","created_at":"2026-07-05T09:20:16.852491+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":16,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01966","citing_title":"Probabilistic Low-Voltage Peak Load Forecasting with Time Series Foundation Models Evaluated on Application-Oriented Metrics","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13024","citing_title":"CausalMoE: A Billion-Scale Multimodal Foundation Model for Granger Causal Discovery with Pattern-Routed Heterogeneous Experts","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11625","citing_title":"TimeRouter: Efficient and Adaptive Routing of Time-Series Foundation Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03530","citing_title":"Towards Intrusion Detection Systems for RPL-based IoT Networks using Foundation Models","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25166","citing_title":"AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2410.04047","citing_title":"TS-Reasoner: Domain-Oriented Time Series Inference Agents for Reasoning and Automated Analysis","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09395","citing_title":"Empowering VLMs for Few-Shot Multimodal Time Series Classification via Tailored Agentic Reasoning","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15549","citing_title":"CTF4Nuclear: Common Task Framework for Nuclear Fission and Fusion Models","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2509.22403","citing_title":"MoveFM-R: Advancing Mobility Foundation Models via Language-driven Semantic Reasoning","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2603.04791","citing_title":"Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10330","citing_title":"Fast Training of Mixture-of-Experts for Time Series Forecasting via Expert Loss Integration","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09395","citing_title":"Empowering VLMs for Few-Shot Multimodal Time Series Classification via Tailored Agentic Reasoning","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22077","citing_title":"Empirical Assessment of Time-Series Foundation Models For Power System Forecasting Applications","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11529","citing_title":"TempusBench: An Evaluation Framework for Time-Series Forecasting","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04475","citing_title":"Discrete Prototypical Memories for Federated Time Series Foundation Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18058","citing_title":"Sonata: A Hybrid World Model for Inertial Kinematics under Clinical Data Scarcity","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB","json":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB.json","graph_json":"https://pith.science/api/pith-number/7HTTNDGSPXUCM6OI6CFJBBJADB/graph.json","events_json":"https://pith.science/api/pith-number/7HTTNDGSPXUCM6OI6CFJBBJADB/events.json","paper":"https://pith.science/paper/7HTTNDGS"},"agent_actions":{"view_html":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB","download_json":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB.json","view_paper":"https://pith.science/paper/7HTTNDGS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.10469&json=true","fetch_graph":"https://pith.science/api/pith-number/7HTTNDGSPXUCM6OI6CFJBBJADB/graph.json","fetch_events":"https://pith.science/api/pith-number/7HTTNDGSPXUCM6OI6CFJBBJADB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB/action/storage_attestation","attest_author":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB/action/author_attestation","sign_citation":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB/action/citation_signature","submit_replication":"https://pith.science/pith/7HTTNDGSPXUCM6OI6CFJBBJADB/action/replication_record"}},"created_at":"2026-07-05T09:20:16.852491+00:00","updated_at":"2026-07-05T09:20:16.852491+00:00"}