{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:QFSIJAERBBF3KRER6WLDN22AES","short_pith_number":"pith:QFSIJAER","schema_version":"1.0","canonical_sha256":"8164848091084bb54491f59636eb40248f5d467966925af5f21a8cfdaef40c8d","source":{"kind":"arxiv","id":"2502.17361","version":2},"attestation_state":"computed","paper":{"title":"A Closer Look at TabPFN v2: Understanding Its Strengths and Extending Its Capabilities","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Han-Jia Ye, Si-Yang Liu, Wei-Lun Chao","submitted_at":"2025-02-24T17:38:42Z","abstract_excerpt":"Tabular datasets are inherently heterogeneous, presenting significant challenges for developing pre-trained foundation models. The recently introduced transformer-based Tabular Prior-data Fitted Network v2 (TabPFN v2) achieves unprecedented in-context learning performance across diverse downstream datasets, marking a pivotal advancement in tabular foundation models. In this paper, we take a closer look at TabPFN v2 to examine how it effectively handles heterogeneity and achieves high predictive accuracy, and to explore how its limitations in high-dimensional, many-category, and large-scale tas"},"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":"2502.17361","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-24T17:38:42Z","cross_cats_sorted":[],"title_canon_sha256":"18684f67c40ec81200bc0a3a7991f62452df5c23db60719c597cd200388ca677","abstract_canon_sha256":"d57266e2a9f403a896c4356098873f0d0567343119d9ed8b0e54bfdc8b291795"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:24.911726Z","signature_b64":"lpm584drBcg9xIjHDibgOH1gTzHjTzjI+GyGRhqVTHcb36qLsW5chQ0SLcHCAMp15s1Yb3LVDpX+Tyr1e7DGDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8164848091084bb54491f59636eb40248f5d467966925af5f21a8cfdaef40c8d","last_reissued_at":"2026-07-05T11:19:24.911245Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:24.911245Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Closer Look at TabPFN v2: Understanding Its Strengths and Extending Its Capabilities","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Han-Jia Ye, Si-Yang Liu, Wei-Lun Chao","submitted_at":"2025-02-24T17:38:42Z","abstract_excerpt":"Tabular datasets are inherently heterogeneous, presenting significant challenges for developing pre-trained foundation models. The recently introduced transformer-based Tabular Prior-data Fitted Network v2 (TabPFN v2) achieves unprecedented in-context learning performance across diverse downstream datasets, marking a pivotal advancement in tabular foundation models. In this paper, we take a closer look at TabPFN v2 to examine how it effectively handles heterogeneity and achieves high predictive accuracy, and to explore how its limitations in high-dimensional, many-category, and large-scale tas"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.17361","kind":"arxiv","version":2},"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/2502.17361/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":"2502.17361","created_at":"2026-07-05T11:19:24.911301+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.17361v2","created_at":"2026-07-05T11:19:24.911301+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.17361","created_at":"2026-07-05T11:19:24.911301+00:00"},{"alias_kind":"pith_short_12","alias_value":"QFSIJAERBBF3","created_at":"2026-07-05T11:19:24.911301+00:00"},{"alias_kind":"pith_short_16","alias_value":"QFSIJAERBBF3KRER","created_at":"2026-07-05T11:19:24.911301+00:00"},{"alias_kind":"pith_short_8","alias_value":"QFSIJAER","created_at":"2026-07-05T11:19:24.911301+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25197","citing_title":"Efficient Adaptive Data Acquisition via Pretrained Belief Representations","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11473","citing_title":"CRUMB: Efficient Prior Fitted Network Inference via Distributionally Matched Context Batching","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29791","citing_title":"What Drives the Inlier-Memorization Effect? A Theory of Outlier Detection via Early Training Dynamics","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30410","citing_title":"Beyond IID: How General Are Tabular Foundation Models, Really?","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20234","citing_title":"TabPFN-MT: A Natively Multitask In-Context Learner for Tabular Data","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20674","citing_title":"Modular Multimodal Classification Without Fine-Tuning: A Simple Compositional Approach","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18147","citing_title":"Foundation Models for Credit Risk Prediction: A Game Changer?","ref_index":118,"is_internal_anchor":false},{"citing_arxiv_id":"2506.02978","citing_title":"On the Robustness of Tabular Foundation Models: Test-Time Attacks and In-Context Defenses","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2506.16791","citing_title":"TabArena: A Living Benchmark for Machine Learning on Tabular Data","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2512.04292","citing_title":"SQuARE: Structured Query & Adaptive Retrieval Engine For Tabular Formats","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04868","citing_title":"Noise Immunity in In-Context Tabular Learning: An Empirical Robustness Analysis of TabPFN's Attention Mechanisms","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11394","citing_title":"Optimizing IoT Intrusion Detection with Tabular Foundation Models for Smart City Forensics","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES","json":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES.json","graph_json":"https://pith.science/api/pith-number/QFSIJAERBBF3KRER6WLDN22AES/graph.json","events_json":"https://pith.science/api/pith-number/QFSIJAERBBF3KRER6WLDN22AES/events.json","paper":"https://pith.science/paper/QFSIJAER"},"agent_actions":{"view_html":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES","download_json":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES.json","view_paper":"https://pith.science/paper/QFSIJAER","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.17361&json=true","fetch_graph":"https://pith.science/api/pith-number/QFSIJAERBBF3KRER6WLDN22AES/graph.json","fetch_events":"https://pith.science/api/pith-number/QFSIJAERBBF3KRER6WLDN22AES/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES/action/storage_attestation","attest_author":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES/action/author_attestation","sign_citation":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES/action/citation_signature","submit_replication":"https://pith.science/pith/QFSIJAERBBF3KRER6WLDN22AES/action/replication_record"}},"created_at":"2026-07-05T11:19:24.911301+00:00","updated_at":"2026-07-05T11:19:24.911301+00:00"}