{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5AC246MXVXOZO43WACG5GZEYOE","short_pith_number":"pith:5AC246MX","schema_version":"1.0","canonical_sha256":"e805ae7997addd977376008dd36498710432576b317d2767902dfeed73edb868","source":{"kind":"arxiv","id":"2606.04876","version":1},"attestation_state":"computed","paper":{"title":"Towards Pretraining Text Encoders for TabPFN","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander Pfefferle, Frank Hutter, Lennart Purucker, Mustafa Tajjar","submitted_at":"2026-06-03T13:38:47Z","abstract_excerpt":"Tabular foundation models, such as TabPFN, achieve strong performance on tabular datasets with numerical and categorical data, but do not natively handle high-cardinality text features. Standard pipelines, therefore, embed text with a language model and compress the resulting vectors with PCA into a small number of scalar features before inputting them into TabPFN. This creates an information bottleneck: most embedding dimensions are discarded, and the compressed representation must then be expanded again by TabPFN's feature encoder. End-to-end alternatives can avoid PCA, but they require larg"},"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":"2606.04876","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-03T13:38:47Z","cross_cats_sorted":[],"title_canon_sha256":"2ab0c60702c94b1a32edba8ccb584a6b171af4bda503f1aaa1b396a0db1412b5","abstract_canon_sha256":"ddb025093c48f91a0b41ee403d40748f16fa435f5c0e7a59d0b47564be5f6edd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:09:52.886335Z","signature_b64":"CCEOEiuWJ7c2w5ZSApfDPc7HpbRQJRB9QUE2xFRKia2+PhOfPN6gfOyGYEcw+WKs8TvTAhBeJHPfrNjVxOz+Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e805ae7997addd977376008dd36498710432576b317d2767902dfeed73edb868","last_reissued_at":"2026-06-04T01:09:52.885477Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:09:52.885477Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Pretraining Text Encoders for TabPFN","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexander Pfefferle, Frank Hutter, Lennart Purucker, Mustafa Tajjar","submitted_at":"2026-06-03T13:38:47Z","abstract_excerpt":"Tabular foundation models, such as TabPFN, achieve strong performance on tabular datasets with numerical and categorical data, but do not natively handle high-cardinality text features. Standard pipelines, therefore, embed text with a language model and compress the resulting vectors with PCA into a small number of scalar features before inputting them into TabPFN. This creates an information bottleneck: most embedding dimensions are discarded, and the compressed representation must then be expanded again by TabPFN's feature encoder. End-to-end alternatives can avoid PCA, but they require larg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.04876","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/2606.04876/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":"2606.04876","created_at":"2026-06-04T01:09:52.885626+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.04876v1","created_at":"2026-06-04T01:09:52.885626+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.04876","created_at":"2026-06-04T01:09:52.885626+00:00"},{"alias_kind":"pith_short_12","alias_value":"5AC246MXVXOZ","created_at":"2026-06-04T01:09:52.885626+00:00"},{"alias_kind":"pith_short_16","alias_value":"5AC246MXVXOZO43W","created_at":"2026-06-04T01:09:52.885626+00:00"},{"alias_kind":"pith_short_8","alias_value":"5AC246MX","created_at":"2026-06-04T01:09:52.885626+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/5AC246MXVXOZO43WACG5GZEYOE","json":"https://pith.science/pith/5AC246MXVXOZO43WACG5GZEYOE.json","graph_json":"https://pith.science/api/pith-number/5AC246MXVXOZO43WACG5GZEYOE/graph.json","events_json":"https://pith.science/api/pith-number/5AC246MXVXOZO43WACG5GZEYOE/events.json","paper":"https://pith.science/paper/5AC246MX"},"agent_actions":{"view_html":"https://pith.science/pith/5AC246MXVXOZO43WACG5GZEYOE","download_json":"https://pith.science/pith/5AC246MXVXOZO43WACG5GZEYOE.json","view_paper":"https://pith.science/paper/5AC246MX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.04876&json=true","fetch_graph":"https://pith.science/api/pith-number/5AC246MXVXOZO43WACG5GZEYOE/graph.json","fetch_events":"https://pith.science/api/pith-number/5AC246MXVXOZO43WACG5GZEYOE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5AC246MXVXOZO43WACG5GZEYOE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5AC246MXVXOZO43WACG5GZEYOE/action/storage_attestation","attest_author":"https://pith.science/pith/5AC246MXVXOZO43WACG5GZEYOE/action/author_attestation","sign_citation":"https://pith.science/pith/5AC246MXVXOZO43WACG5GZEYOE/action/citation_signature","submit_replication":"https://pith.science/pith/5AC246MXVXOZO43WACG5GZEYOE/action/replication_record"}},"created_at":"2026-06-04T01:09:52.885626+00:00","updated_at":"2026-06-04T01:09:52.885626+00:00"}