{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:ZWYL2XTTLZP7C7SJXLRYXDXL7Q","short_pith_number":"pith:ZWYL2XTT","schema_version":"1.0","canonical_sha256":"cdb0bd5e735e5ff17e49bae38b8eebfc128571cd8d9e194e25807fddafb5333e","source":{"kind":"arxiv","id":"2106.01342","version":1},"attestation_state":"computed","paper":{"title":"SAINT: Improved Neural Networks for Tabular Data via Row Attention and Contrastive Pre-Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Avi Schwarzschild, C. Bayan Bruss, Gowthami Somepalli, Micah Goldblum, Tom Goldstein","submitted_at":"2021-06-02T17:51:05Z","abstract_excerpt":"Tabular data underpins numerous high-impact applications of machine learning from fraud detection to genomics and healthcare. Classical approaches to solving tabular problems, such as gradient boosting and random forests, are widely used by practitioners. However, recent deep learning methods have achieved a degree of performance competitive with popular techniques. We devise a hybrid deep learning approach to solving tabular data problems. Our method, SAINT, performs attention over both rows and columns, and it includes an enhanced embedding method. We also study a new contrastive self-superv"},"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":"2106.01342","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-02T17:51:05Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"48824777b04b53182b8063b2b81d1b6ebe0ea26b1662e465a494a323b6db6c71","abstract_canon_sha256":"34b56a162b06b5aa2ef600b2adc52ec345a96270e395616df5a58a2c44d70104"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:45:46.703695Z","signature_b64":"tRcIcl0cj4KR6p3JtE1BfIRStCJzXXPIkpT6nVtC8n7APkUhnel/ZSZ3RjuVHTOoAX8SAW0mq3trXc+KZDVgDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cdb0bd5e735e5ff17e49bae38b8eebfc128571cd8d9e194e25807fddafb5333e","last_reissued_at":"2026-07-05T02:45:46.703244Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:45:46.703244Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SAINT: Improved Neural Networks for Tabular Data via Row Attention and Contrastive Pre-Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Avi Schwarzschild, C. Bayan Bruss, Gowthami Somepalli, Micah Goldblum, Tom Goldstein","submitted_at":"2021-06-02T17:51:05Z","abstract_excerpt":"Tabular data underpins numerous high-impact applications of machine learning from fraud detection to genomics and healthcare. Classical approaches to solving tabular problems, such as gradient boosting and random forests, are widely used by practitioners. However, recent deep learning methods have achieved a degree of performance competitive with popular techniques. We devise a hybrid deep learning approach to solving tabular data problems. Our method, SAINT, performs attention over both rows and columns, and it includes an enhanced embedding method. We also study a new contrastive self-superv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.01342","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/2106.01342/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":"2106.01342","created_at":"2026-07-05T02:45:46.703302+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.01342v1","created_at":"2026-07-05T02:45:46.703302+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.01342","created_at":"2026-07-05T02:45:46.703302+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZWYL2XTTLZP7","created_at":"2026-07-05T02:45:46.703302+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZWYL2XTTLZP7C7SJ","created_at":"2026-07-05T02:45:46.703302+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZWYL2XTT","created_at":"2026-07-05T02:45:46.703302+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":19,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.09323","citing_title":"TRL-Bench: Standardizing Cross-Paradigm Representation-Level Evaluation of Tabular Encoders","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04445","citing_title":"RowNet: A Memory Transformer for Tabular Regression","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01189","citing_title":"The Case for Model Science: Verify, Explore, Steer, Refine","ref_index":253,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04363","citing_title":"Mitigating Label Shift in Tabular In-Context Learning via Test-Time Posterior Adjustment","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23241","citing_title":"RelPrism: A Multi-Faceted Pre-training Framework with Self-Generated Tasks for Relational Databases","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2603.16513","citing_title":"FEAT: A Linear-Complexity Foundation Model for Extremely Large Structured Data","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20234","citing_title":"TabPFN-MT: A Natively Multitask In-Context Learner for Tabular Data","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18147","citing_title":"Foundation Models for Credit Risk Prediction: A Game Changer?","ref_index":159,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19014","citing_title":"SAGA: A Sequence-Adaptive Generative Architecture for Multi-Horizon Probabilistic Forecasting with Adaptive Temporal Conformal Prediction","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16085","citing_title":"Towards Foundation Models for Relational Databases with Language Models and Graph Neural Networks","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2509.11449","citing_title":"Tabular Data with Class Imbalance: Predicting Electric Vehicle Crash Severity with Pretrained Transformers (TabPFN) and Mamba-Based Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2602.20223","citing_title":"MultiModalPFN: Extending Prior-Data Fitted Networks for Multimodal Tabular Learning","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2603.21236","citing_title":"Posterior-Calibrated Causal Circuits in Variational Autoencoders: Why Image-Domain Interpretability Fails on Tabular Data","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2207.01848","citing_title":"TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10337","citing_title":"Integrating SAINT with Tree-Based Models: A Case Study in Employee Attrition Prediction","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08649","citing_title":"PRAGMA: Revolut Foundation Model","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05857","citing_title":"Weight-Informed Self-Explaining Clustering for Mixed-Type Tabular Data","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05635","citing_title":"From Uniform to Learned Knots: A Study of Spline-Based Numerical Encodings for Tabular Deep Learning","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04363","citing_title":"Mitigating Label Shift in Tabular In-Context Learning via Test-Time Posterior Adjustment","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q","json":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q.json","graph_json":"https://pith.science/api/pith-number/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/graph.json","events_json":"https://pith.science/api/pith-number/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/events.json","paper":"https://pith.science/paper/ZWYL2XTT"},"agent_actions":{"view_html":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q","download_json":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q.json","view_paper":"https://pith.science/paper/ZWYL2XTT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.01342&json=true","fetch_graph":"https://pith.science/api/pith-number/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/graph.json","fetch_events":"https://pith.science/api/pith-number/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/action/storage_attestation","attest_author":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/action/author_attestation","sign_citation":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/action/citation_signature","submit_replication":"https://pith.science/pith/ZWYL2XTTLZP7C7SJXLRYXDXL7Q/action/replication_record"}},"created_at":"2026-07-05T02:45:46.703302+00:00","updated_at":"2026-07-05T02:45:46.703302+00:00"}