{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:DAJLSYUWH5NXQRS5IFOXJJ5BG3","short_pith_number":"pith:DAJLSYUW","schema_version":"1.0","canonical_sha256":"1812b962963f5b78465d415d74a7a136f2b2baa0d83255c4504f5819b81239a2","source":{"kind":"arxiv","id":"2404.11757","version":1},"attestation_state":"computed","paper":{"title":"Language Models Still Struggle to Zero-shot Reason about Time Series","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Mike A. Merrill, Mingtian Tan, Tim Althoff, Tom Hartvigsen, Vinayak Gupta","submitted_at":"2024-04-17T21:27:33Z","abstract_excerpt":"Time series are critical for decision-making in fields like finance and healthcare. Their importance has driven a recent influx of works passing time series into language models, leading to non-trivial forecasting on some datasets. But it remains unknown whether non-trivial forecasting implies that language models can reason about time series. To address this gap, we generate a first-of-its-kind evaluation framework for time series reasoning, including formal tasks and a corresponding dataset of multi-scale time series paired with text captions across ten domains. Using these data, we probe wh"},"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":"2404.11757","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-17T21:27:33Z","cross_cats_sorted":[],"title_canon_sha256":"e6cc38374749ceb84bdc33f21754dfc360ec760af17c83ca4cafa5112545ded5","abstract_canon_sha256":"000918cede8b23ca1c2647d70c1dd7f835a186daeb9bbede24957fe277ddacc9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:09:15.395582Z","signature_b64":"YBEn1eJ3Ss+Jiyhj30X47ToYXNmxGPfhw1LCpCV3J84U+p8bdpDxrjknN2mJ1WtvxJ9NSSM7ccabT4OJ12IiBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1812b962963f5b78465d415d74a7a136f2b2baa0d83255c4504f5819b81239a2","last_reissued_at":"2026-07-05T08:09:15.395059Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:09:15.395059Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Language Models Still Struggle to Zero-shot Reason about Time Series","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Mike A. Merrill, Mingtian Tan, Tim Althoff, Tom Hartvigsen, Vinayak Gupta","submitted_at":"2024-04-17T21:27:33Z","abstract_excerpt":"Time series are critical for decision-making in fields like finance and healthcare. Their importance has driven a recent influx of works passing time series into language models, leading to non-trivial forecasting on some datasets. But it remains unknown whether non-trivial forecasting implies that language models can reason about time series. To address this gap, we generate a first-of-its-kind evaluation framework for time series reasoning, including formal tasks and a corresponding dataset of multi-scale time series paired with text captions across ten domains. Using these data, we probe wh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.11757","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/2404.11757/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":"2404.11757","created_at":"2026-07-05T08:09:15.395126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.11757v1","created_at":"2026-07-05T08:09:15.395126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.11757","created_at":"2026-07-05T08:09:15.395126+00:00"},{"alias_kind":"pith_short_12","alias_value":"DAJLSYUWH5NX","created_at":"2026-07-05T08:09:15.395126+00:00"},{"alias_kind":"pith_short_16","alias_value":"DAJLSYUWH5NXQRS5","created_at":"2026-07-05T08:09:15.395126+00:00"},{"alias_kind":"pith_short_8","alias_value":"DAJLSYUW","created_at":"2026-07-05T08:09:15.395126+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.09741","citing_title":"FoNE: Precise Single-Token Number Embeddings via Fourier Features","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20268","citing_title":"Chronicle: A Multimodal Foundation Model for Joint Language and Time Series Understanding","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2510.07432","citing_title":"TS-Agent: Understanding and Reasoning Over Raw Time Series via Iterative Insight Gathering","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2511.01101","citing_title":"TSVer: A Benchmark for Fact Verification Against Time-Series Evidence","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2603.12451","citing_title":"Overcoming the Modality Gap in Context-Aided Forecasting","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10291","citing_title":"TimeSeriesExamAgent: Creating Time Series Reasoning Benchmarks at Scale","ref_index":29,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3","json":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3.json","graph_json":"https://pith.science/api/pith-number/DAJLSYUWH5NXQRS5IFOXJJ5BG3/graph.json","events_json":"https://pith.science/api/pith-number/DAJLSYUWH5NXQRS5IFOXJJ5BG3/events.json","paper":"https://pith.science/paper/DAJLSYUW"},"agent_actions":{"view_html":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3","download_json":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3.json","view_paper":"https://pith.science/paper/DAJLSYUW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.11757&json=true","fetch_graph":"https://pith.science/api/pith-number/DAJLSYUWH5NXQRS5IFOXJJ5BG3/graph.json","fetch_events":"https://pith.science/api/pith-number/DAJLSYUWH5NXQRS5IFOXJJ5BG3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3/action/storage_attestation","attest_author":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3/action/author_attestation","sign_citation":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3/action/citation_signature","submit_replication":"https://pith.science/pith/DAJLSYUWH5NXQRS5IFOXJJ5BG3/action/replication_record"}},"created_at":"2026-07-05T08:09:15.395126+00:00","updated_at":"2026-07-05T08:09:15.395126+00:00"}