{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "ekambaram-et-al-tiny-time-mixers-2024", "title": "Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting", "summary": "Adapted lightweight mixer blocks for pretrained forecasting across heterogeneous temporal resolutions using adaptive patching, resolution sampling/prefixes and channel-independent pretraining followed by channel/exogenous mixing during fine-tuning.", "boundary": "TSMixer backbone is inherited. Transfer claims concern specified forecasting datasets, not universal time-series guarantees.", "area": "Efficient time-series foundation models", "topic": "Artificial intelligence", "year": 2024, "lane": "HUMAN", "result": "EMPIRICAL", "package_id": "ekambaram-et-al-tiny-time-mixers-2024", "target": { "record_id": "urn:uuid:fb50490c-d05c-4f6f-a27f-6e0c5a64de24", "version": 1, "content_hash": "sha256:c7b473e93e4b0e387873c0c51daa4aababde7d351764a7dc10461b34653965bb" }, "humans": [ "Vijay Ekambaram", "Arindam Jati", "Pankaj Dayama", "Sumanta Mukherjee", "Nam H. Nguyen", "Wesley M. Gifford", "Chandra Reddy", "Jayant Kalagnanam" ], "authors_complete": true, "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "cur20261015-neurips-ttm-final", "title": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" }, { "id": "cur20261015-arxiv-2401-03955v1", "title": "TTMs: Fast Multi-level Tiny Time Mixers for Improved Zero-shot and Few-shot Forecasting of Multivariate Time Series", "url": "https://arxiv.org/abs/2401.03955v1" }, { "id": "cur20261015-support-ekambaram-et-al-tiny-time-mixers-2024", "title": "Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting — official research provenance", "url": "https://research.ibm.com/publications/tiny-time-mixers-ttms-fast-pre-trained-models-for-enhanced-zerofew-shot-forecasting-of-multivariate-time-series--1" } ], "points": "65000", "published_grade": { "id": "ekambaram-et-al-tiny-time-mixers-2024", "entry_id": "ekambaram-et-al-tiny-time-mixers-2024", "band": "B", "modifier": "+", "target": { "record_id": "urn:uuid:fb50490c-d05c-4f6f-a27f-6e0c5a64de24", "version": 1, "content_hash": "sha256:c7b473e93e4b0e387873c0c51daa4aababde7d351764a7dc10461b34653965bb" }, "what_changed": "Adapted lightweight mixer blocks for pretrained forecasting across heterogeneous temporal resolutions using adaptive patching, resolution sampling/prefixes and channel-independent pretraining followed by channel/exogenous mixing during fine-tuning.", "scope": "Adapted lightweight mixer blocks for pretrained forecasting across heterogeneous temporal resolutions using adaptive patching, resolution sampling/prefixes and channel-independent pretraining followed by channel/exogenous mixing during fine-tuning. TSMixer backbone is inherited. Transfer claims concern specified forecasting datasets, not universal time-series guarantees.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:fb50490c-d05c-4f6f-a27f-6e0c5a64de24", "version": 1, "content_hash": "sha256:c7b473e93e4b0e387873c0c51daa4aababde7d351764a7dc10461b34653965bb" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-10-16T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Vijay Ekambaram", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Arindam Jati", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Pankaj Dayama", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Sumanta Mukherjee", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Nam H. Nguyen", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Wesley M. Gifford", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Chandra Reddy", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Jayant Kalagnanam", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "discoverer" ], "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" } }, { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-1", "kind": "source_reference", "source": { "citation": "TTMs: Fast Multi-level Tiny Time Mixers for Improved Zero-shot and Few-shot Forecasting of Multivariate Time Series", "url": "https://arxiv.org/abs/2401.03955v1" } }, { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-2", "kind": "source_reference", "source": { "citation": "Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting — official research provenance", "url": "https://research.ibm.com/publications/tiny-time-mixers-ttms-fast-pre-trained-models-for-enhanced-zerofew-shot-forecasting-of-multivariate-time-series--1" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "TSMixer backbone is inherited. Transfer claims concern specified forecasting datasets, not universal time-series guarantees.", "domain": "Efficient time-series foundation models" }, "sources": [ { "citation": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf" }, { "citation": "TTMs: Fast Multi-level Tiny Time Mixers for Improved Zero-shot and Few-shot Forecasting of Multivariate Time Series", "url": "https://arxiv.org/abs/2401.03955v1" }, { "citation": "Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting — official research provenance", "url": "https://research.ibm.com/publications/tiny-time-mixers-ttms-fast-pre-trained-models-for-enhanced-zerofew-shot-forecasting-of-multivariate-time-series--1" } ], "statement": "# Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting\n\nAdapted lightweight mixer blocks for pretrained forecasting across heterogeneous temporal resolutions using adaptive patching, resolution sampling/prefixes and channel-independent pretraining followed by channel/exogenous mixing during fine-tuning.\n\n## Scientific scope\n\nTSMixer backbone is inherited. Transfer claims concern specified forecasting datasets, not universal time-series guarantees.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nVijay Ekambaram; Arindam Jati; Pankaj Dayama; Sumanta Mukherjee; Nam H. Nguyen; Wesley M. Gifford; Chandra Reddy; Jayant Kalagnanam\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series](https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf)\n- [TTMs: Fast Multi-level Tiny Time Mixers for Improved Zero-shot and Few-shot Forecasting of Multivariate Time Series](https://arxiv.org/abs/2401.03955v1)\n- [Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting — official research provenance](https://research.ibm.com/publications/tiny-time-mixers-ttms-fast-pre-trained-models-for-enhanced-zerofew-shot-forecasting-of-multivariate-time-series--1)\n", "summary": "Adapted lightweight mixer blocks for pretrained forecasting across heterogeneous temporal resolutions using adaptive patching, resolution sampling/prefixes and channel-independent pretraining followed by channel/exogenous mixing during fine-tuning.", "title": "Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting" }, "record_id": "urn:uuid:fb50490c-d05c-4f6f-a27f-6e0c5a64de24", "version": 1, "source_ref": { "record_id": "urn:uuid:fb50490c-d05c-4f6f-a27f-6e0c5a64de24", "version": 1, "content_hash": "sha256:c7b473e93e4b0e387873c0c51daa4aababde7d351764a7dc10461b34653965bb" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:ea9f8e9b4a377151d7629eb3cb6eb3082daafe1eb7123bf0556c34023e1d5067" }, "target": { "record_id": "urn:uuid:fb50490c-d05c-4f6f-a27f-6e0c5a64de24", "version": 1, "content_hash": "sha256:c7b473e93e4b0e387873c0c51daa4aababde7d351764a7dc10461b34653965bb" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "ekambaram-et-al-tiny-time-mixers-2024", "entry_id": "ekambaram-et-al-tiny-time-mixers-2024", "revision": 1, "title": "Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting", "context": "Efficient time-series foundation models", "track": "Human-designed research method or resource", "band": "B", "modifier": "+", "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-10-16", "evidence_cutoff": "2026-10-16", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Adapted lightweight mixer blocks for pretrained forecasting across heterogeneous temporal resolutions using adaptive patching, resolution sampling/prefixes and channel-independent pretraining followed by channel/exogenous mixing during fine-tuning.", "scope": "Adapted lightweight mixer blocks for pretrained forecasting across heterogeneous temporal resolutions using adaptive patching, resolution sampling/prefixes and channel-independent pretraining followed by channel/exogenous mixing during fine-tuning. TSMixer backbone is inherited. Transfer claims concern specified forecasting datasets, not universal time-series guarantees.", "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. TSMixer backbone is inherited. Transfer claims concern specified forecasting datasets, not universal time-series guarantees.", "references": [ { "title": "Tiny Time Mixers (TTMs): Fast Pre-trained Models for Enhanced Zero/Few-shot Forecasting of Multivariate Time Series", "url": "https://proceedings.nips.cc/paper_files/paper/2024/file/874a4d89f2d04b4bcf9a2c19545cf040-Paper-Conference.pdf", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." }, { "title": "TTMs: Fast Multi-level Tiny Time Mixers for Improved Zero-shot and Few-shot Forecasting of Multivariate Time Series", "url": "https://arxiv.org/abs/2401.03955v1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." }, { "title": "Ekambaram et al.: Tiny Time Mixers for transferable time-series forecasting — official research provenance", "url": "https://research.ibm.com/publications/tiny-time-mixers-ttms-fast-pre-trained-models-for-enhanced-zerofew-shot-forecasting-of-multivariate-time-series--1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." } ], "roles": [ { "who": "Source-listed authors and technical contributors", "credit": "SCIENCE", "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities." } ], "target": { "record_id": "urn:uuid:fb50490c-d05c-4f6f-a27f-6e0c5a64de24", "version": 1, "content_hash": "sha256:c7b473e93e4b0e387873c0c51daa4aababde7d351764a7dc10461b34653965bb" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "65000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }