{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZE2ZV2RMXCKR4ICAMYSYD7SKBX","short_pith_number":"pith:ZE2ZV2RM","schema_version":"1.0","canonical_sha256":"c9359aea2cb8951e2040662581fe4a0def9e122482ca8cc010c8faeb24f831e9","source":{"kind":"arxiv","id":"2506.05515","version":2},"attestation_state":"computed","paper":{"title":"Winner-takes-all for Multivariate Probabilistic Time Series Forecasting","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Adrien Cort\\'es, R\\'emi Rehm, Victor Letzelter","submitted_at":"2025-06-05T18:56:14Z","abstract_excerpt":"We introduce TimeMCL, a method leveraging the Multiple Choice Learning (MCL) paradigm to forecast multiple plausible time series futures. Our approach employs a neural network with multiple heads and utilizes the Winner-Takes-All (WTA) loss to promote diversity among predictions. MCL has recently gained attention due to its simplicity and ability to address ill-posed and ambiguous tasks. We propose an adaptation of this framework for time-series forecasting, presenting it as an efficient method to predict diverse futures, which we relate to its implicit quantization objective. We provide insig"},"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":"2506.05515","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-05T18:56:14Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"b8f0854c79afc6c9f8c2fb1a961582d659463bc3fd85488d0330e0f943969935","abstract_canon_sha256":"45390f8e8b02386efbbffdcbe856cd710ef5ef1a87eb5b6c11d6aac18cbac6a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:52:02.572827Z","signature_b64":"eE6EUFnCqI5SkCXLnZzk6bhKtwkO/hhLwS6Q3h8yhN4MwVor7paNpawaOdYXl3utz/8oznambL3v7FMZBRiVAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9359aea2cb8951e2040662581fe4a0def9e122482ca8cc010c8faeb24f831e9","last_reissued_at":"2026-07-05T11:52:02.572379Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:52:02.572379Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Winner-takes-all for Multivariate Probabilistic Time Series Forecasting","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Adrien Cort\\'es, R\\'emi Rehm, Victor Letzelter","submitted_at":"2025-06-05T18:56:14Z","abstract_excerpt":"We introduce TimeMCL, a method leveraging the Multiple Choice Learning (MCL) paradigm to forecast multiple plausible time series futures. Our approach employs a neural network with multiple heads and utilizes the Winner-Takes-All (WTA) loss to promote diversity among predictions. MCL has recently gained attention due to its simplicity and ability to address ill-posed and ambiguous tasks. We propose an adaptation of this framework for time-series forecasting, presenting it as an efficient method to predict diverse futures, which we relate to its implicit quantization objective. We provide insig"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05515","kind":"arxiv","version":2},"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/2506.05515/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":"2506.05515","created_at":"2026-07-05T11:52:02.572438+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.05515v2","created_at":"2026-07-05T11:52:02.572438+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05515","created_at":"2026-07-05T11:52:02.572438+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZE2ZV2RMXCKR","created_at":"2026-07-05T11:52:02.572438+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZE2ZV2RMXCKR4ICA","created_at":"2026-07-05T11:52:02.572438+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZE2ZV2RM","created_at":"2026-07-05T11:52:02.572438+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.28867","citing_title":"PrismFlow: Residual Dynamics for Flow Matching in Time-Series Generation","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23402","citing_title":"Parametric Prior Mapping Framework for Non-stationary Probabilistic Time Series Forecasting","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2509.25914","citing_title":"ReNF: Rethinking the Design of Neural Long-Term Time Series Forecasters","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX","json":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX.json","graph_json":"https://pith.science/api/pith-number/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/graph.json","events_json":"https://pith.science/api/pith-number/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/events.json","paper":"https://pith.science/paper/ZE2ZV2RM"},"agent_actions":{"view_html":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX","download_json":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX.json","view_paper":"https://pith.science/paper/ZE2ZV2RM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.05515&json=true","fetch_graph":"https://pith.science/api/pith-number/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/graph.json","fetch_events":"https://pith.science/api/pith-number/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/action/storage_attestation","attest_author":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/action/author_attestation","sign_citation":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/action/citation_signature","submit_replication":"https://pith.science/pith/ZE2ZV2RMXCKR4ICAMYSYD7SKBX/action/replication_record"}},"created_at":"2026-07-05T11:52:02.572438+00:00","updated_at":"2026-07-05T11:52:02.572438+00:00"}