{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CS3LK5FDPAST2LYK3N3LB36MBY","short_pith_number":"pith:CS3LK5FD","schema_version":"1.0","canonical_sha256":"14b6b574a378253d2f0adb76b0efcc0e2f2071011e509b87552876a62023779e","source":{"kind":"arxiv","id":"2605.21241","version":1},"attestation_state":"computed","paper":{"title":"Divide and Contrast: Learning Robust Temporal Features without Augmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Abdul-Kazeem Shamba, Gavin Taylor, Kerstin Bach","submitted_at":"2026-05-20T14:31:44Z","abstract_excerpt":"Self-supervised learning for time-series representation aims to reduce reliance on labeled data while maintaining strong downstream performance, yet many existing approaches incur high computational costs or rely on assumptions that do not hold across diverse temporal dynamics. In this work, we introduce Divide and Contrast (Di-COT), an unsupervised framework that avoids data augmentation and multiple encoder passes by contrasting informative substructures within a window rather than individual timesteps. Di-COT stochastically partitions each window into a small number of overlapping sub-block"},"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":"2605.21241","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-20T14:31:44Z","cross_cats_sorted":[],"title_canon_sha256":"3ae8fbde70c1e513c77805b9c512b2c5828e95a0e1619e4214a383e330b988d8","abstract_canon_sha256":"33536bcb78f05d0460013862432479bacbccc004d46f10d57daf11232e2ea972"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:05:45.373146Z","signature_b64":"vhewyfSaYcDLPriH+hSWxT8MRVT4dyUXHi3mjsNtJLk/u2+xMYkB8RT5+fd0Eez56KGN90vuUqDaIAwaFFtSDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"14b6b574a378253d2f0adb76b0efcc0e2f2071011e509b87552876a62023779e","last_reissued_at":"2026-05-21T01:05:45.372475Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:05:45.372475Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Divide and Contrast: Learning Robust Temporal Features without Augmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Abdul-Kazeem Shamba, Gavin Taylor, Kerstin Bach","submitted_at":"2026-05-20T14:31:44Z","abstract_excerpt":"Self-supervised learning for time-series representation aims to reduce reliance on labeled data while maintaining strong downstream performance, yet many existing approaches incur high computational costs or rely on assumptions that do not hold across diverse temporal dynamics. In this work, we introduce Divide and Contrast (Di-COT), an unsupervised framework that avoids data augmentation and multiple encoder passes by contrasting informative substructures within a window rather than individual timesteps. Di-COT stochastically partitions each window into a small number of overlapping sub-block"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21241","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/2605.21241/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":"2605.21241","created_at":"2026-05-21T01:05:45.372574+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.21241v1","created_at":"2026-05-21T01:05:45.372574+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21241","created_at":"2026-05-21T01:05:45.372574+00:00"},{"alias_kind":"pith_short_12","alias_value":"CS3LK5FDPAST","created_at":"2026-05-21T01:05:45.372574+00:00"},{"alias_kind":"pith_short_16","alias_value":"CS3LK5FDPAST2LYK","created_at":"2026-05-21T01:05:45.372574+00:00"},{"alias_kind":"pith_short_8","alias_value":"CS3LK5FD","created_at":"2026-05-21T01:05:45.372574+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY","json":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY.json","graph_json":"https://pith.science/api/pith-number/CS3LK5FDPAST2LYK3N3LB36MBY/graph.json","events_json":"https://pith.science/api/pith-number/CS3LK5FDPAST2LYK3N3LB36MBY/events.json","paper":"https://pith.science/paper/CS3LK5FD"},"agent_actions":{"view_html":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY","download_json":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY.json","view_paper":"https://pith.science/paper/CS3LK5FD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.21241&json=true","fetch_graph":"https://pith.science/api/pith-number/CS3LK5FDPAST2LYK3N3LB36MBY/graph.json","fetch_events":"https://pith.science/api/pith-number/CS3LK5FDPAST2LYK3N3LB36MBY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY/action/storage_attestation","attest_author":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY/action/author_attestation","sign_citation":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY/action/citation_signature","submit_replication":"https://pith.science/pith/CS3LK5FDPAST2LYK3N3LB36MBY/action/replication_record"}},"created_at":"2026-05-21T01:05:45.372574+00:00","updated_at":"2026-05-21T01:05:45.372574+00:00"}