{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KOMDX7IIBB5XNTTWEXJSNVX2WG","short_pith_number":"pith:KOMDX7II","schema_version":"1.0","canonical_sha256":"53983bfd08087b76ce7625d326d6fab1ab314653cf136ec3bbd1410f05816da3","source":{"kind":"arxiv","id":"2410.24210","version":3},"attestation_state":"computed","paper":{"title":"TabM: Advancing Tabular Deep Learning with Parameter-Efficient Ensembling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Akim Kotelnikov, Artem Babenko, Yury Gorishniy","submitted_at":"2024-10-31T17:58:41Z","abstract_excerpt":"Deep learning architectures for supervised learning on tabular data range from simple multilayer perceptrons (MLP) to sophisticated Transformers and retrieval-augmented methods. This study highlights a major, yet so far overlooked opportunity for designing substantially better MLP-based tabular architectures. Namely, our new model TabM relies on efficient ensembling, where one TabM efficiently imitates an ensemble of MLPs and produces multiple predictions per object. Compared to a traditional deep ensemble, in TabM, the underlying implicit MLPs are trained simultaneously, and (by default) shar"},"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":"2410.24210","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-31T17:58:41Z","cross_cats_sorted":[],"title_canon_sha256":"9040e96157aeee94c646d13c08fd2a3427a72c0c5333d6c8229a0fbdfbeac7c6","abstract_canon_sha256":"34c10a4e0f4ba551445ad2037caa6c289df5a70224b258682e5ca57daf1e35e7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:16:14.724728Z","signature_b64":"0i076+u2xHXGlUrXHgGg2zJGeWaQQ6mg6TkSZFZRObNHxsmuPqIF5UkYHiNG+UURGAsIGzPMqk5dS2o2NoghDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"53983bfd08087b76ce7625d326d6fab1ab314653cf136ec3bbd1410f05816da3","last_reissued_at":"2026-07-05T10:16:14.724196Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:16:14.724196Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TabM: Advancing Tabular Deep Learning with Parameter-Efficient Ensembling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Akim Kotelnikov, Artem Babenko, Yury Gorishniy","submitted_at":"2024-10-31T17:58:41Z","abstract_excerpt":"Deep learning architectures for supervised learning on tabular data range from simple multilayer perceptrons (MLP) to sophisticated Transformers and retrieval-augmented methods. This study highlights a major, yet so far overlooked opportunity for designing substantially better MLP-based tabular architectures. Namely, our new model TabM relies on efficient ensembling, where one TabM efficiently imitates an ensemble of MLPs and produces multiple predictions per object. Compared to a traditional deep ensemble, in TabM, the underlying implicit MLPs are trained simultaneously, and (by default) shar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.24210","kind":"arxiv","version":3},"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/2410.24210/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":"2410.24210","created_at":"2026-07-05T10:16:14.724268+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.24210v3","created_at":"2026-07-05T10:16:14.724268+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.24210","created_at":"2026-07-05T10:16:14.724268+00:00"},{"alias_kind":"pith_short_12","alias_value":"KOMDX7IIBB5X","created_at":"2026-07-05T10:16:14.724268+00:00"},{"alias_kind":"pith_short_16","alias_value":"KOMDX7IIBB5XNTTW","created_at":"2026-07-05T10:16:14.724268+00:00"},{"alias_kind":"pith_short_8","alias_value":"KOMDX7II","created_at":"2026-07-05T10:16:14.724268+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02384","citing_title":"TabPrep: Closing the Feature Engineering Gap in Tabular Benchmarks","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24417","citing_title":"LLMTabBench: Evaluating LLMs on Binary Tabular Classification From Zero to Few Shots","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30410","citing_title":"Beyond IID: How General Are Tabular Foundation Models, Really?","ref_index":87,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28554","citing_title":"High Performance, Low Reliability: Uncertainty Benchmarking for Tabular Foundation Models","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28089","citing_title":"BuddyBench: A Privacy-Constrained Multi-Task Benchmark for Pediatric Social-Communication Personalization","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22644","citing_title":"Why SGD is not Brownian Motion: A New Perspective on Stochastic Dynamics","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2509.21465","citing_title":"Talking Trees: Reasoning-Assisted Induction of Decision Trees for Tabular Data","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2502.05564","citing_title":"TabICL: A Tabular Foundation Model for In-Context Learning on Large Data","ref_index":194,"is_internal_anchor":false},{"citing_arxiv_id":"2506.16791","citing_title":"TabArena: A Living Benchmark for Machine Learning on Tabular Data","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2508.10053","citing_title":"xRFM: Accurate, scalable, and interpretable feature learning models for tabular data","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2512.13040","citing_title":"Understanding Structured Financial Data with LLMs: A Case Study on Fraud Detection","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06047","citing_title":"TFM-Retouche: A Lightweight Input-Space Adapter for Tabular Foundation Models","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10616","citing_title":"MulTaBench: Benchmarking Multimodal Tabular Learning with Text and Image","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06047","citing_title":"TFM-Retouche: A Lightweight Input-Space Adapter for Tabular Foundation Models","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02003","citing_title":"RamanBench: A Large-Scale Benchmark for Machine Learning on Raman Spectroscopy","ref_index":65,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG","json":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG.json","graph_json":"https://pith.science/api/pith-number/KOMDX7IIBB5XNTTWEXJSNVX2WG/graph.json","events_json":"https://pith.science/api/pith-number/KOMDX7IIBB5XNTTWEXJSNVX2WG/events.json","paper":"https://pith.science/paper/KOMDX7II"},"agent_actions":{"view_html":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG","download_json":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG.json","view_paper":"https://pith.science/paper/KOMDX7II","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.24210&json=true","fetch_graph":"https://pith.science/api/pith-number/KOMDX7IIBB5XNTTWEXJSNVX2WG/graph.json","fetch_events":"https://pith.science/api/pith-number/KOMDX7IIBB5XNTTWEXJSNVX2WG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG/action/storage_attestation","attest_author":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG/action/author_attestation","sign_citation":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG/action/citation_signature","submit_replication":"https://pith.science/pith/KOMDX7IIBB5XNTTWEXJSNVX2WG/action/replication_record"}},"created_at":"2026-07-05T10:16:14.724268+00:00","updated_at":"2026-07-05T10:16:14.724268+00:00"}