{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IS3K73OG3TDWAPGXZ4T7ZGC4HJ","short_pith_number":"pith:IS3K73OG","schema_version":"1.0","canonical_sha256":"44b6afedc6dcc7603cd7cf27fc985c3a48a7a980a2d1cf0cd62ecc6aed5a3c73","source":{"kind":"arxiv","id":"2502.09844","version":2},"attestation_state":"computed","paper":{"title":"Solving Empirical Bayes via Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Anzo Teh, Mark Jabbour, Yury Polyanskiy","submitted_at":"2025-02-14T01:06:15Z","abstract_excerpt":"This work applies modern AI tools (transformers) to solving one of the oldest statistical problems: Poisson means under empirical Bayes (Poisson-EB) setting. In Poisson-EB a high-dimensional mean vector $\\theta$ (with iid coordinates sampled from an unknown prior $\\pi$) is estimated on the basis of $X=\\mathrm{Poisson}(\\theta)$. A transformer model is pre-trained on a set of synthetically generated pairs $(X,\\theta)$ and learns to do in-context learning (ICL) by adapting to unknown $\\pi$. Theoretically, we show that a sufficiently wide transformer can achieve vanishing regret with respect to an"},"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":"2502.09844","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-14T01:06:15Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"752a029299db60c386844331ee1f8bdc723f91024d2e74105f0f2777d302bc29","abstract_canon_sha256":"42182050d718d9866b5ea1a5ef8fda8c1be2006c0c503e9f4ce28dd38d98cc76"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:52.427383Z","signature_b64":"dj0gwx35M4DaCLEehannOh86WMkicJFFc0Tl8Piv7F2qJPbxSBEIg00eFLHY3BhSruNFBt4gFi4Yx/vVVzeOBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"44b6afedc6dcc7603cd7cf27fc985c3a48a7a980a2d1cf0cd62ecc6aed5a3c73","last_reissued_at":"2026-07-05T11:10:52.426865Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:52.426865Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Solving Empirical Bayes via Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Anzo Teh, Mark Jabbour, Yury Polyanskiy","submitted_at":"2025-02-14T01:06:15Z","abstract_excerpt":"This work applies modern AI tools (transformers) to solving one of the oldest statistical problems: Poisson means under empirical Bayes (Poisson-EB) setting. In Poisson-EB a high-dimensional mean vector $\\theta$ (with iid coordinates sampled from an unknown prior $\\pi$) is estimated on the basis of $X=\\mathrm{Poisson}(\\theta)$. A transformer model is pre-trained on a set of synthetically generated pairs $(X,\\theta)$ and learns to do in-context learning (ICL) by adapting to unknown $\\pi$. Theoretically, we show that a sufficiently wide transformer can achieve vanishing regret with respect to an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09844","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/2502.09844/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":"2502.09844","created_at":"2026-07-05T11:10:52.426914+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.09844v2","created_at":"2026-07-05T11:10:52.426914+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09844","created_at":"2026-07-05T11:10:52.426914+00:00"},{"alias_kind":"pith_short_12","alias_value":"IS3K73OG3TDW","created_at":"2026-07-05T11:10:52.426914+00:00"},{"alias_kind":"pith_short_16","alias_value":"IS3K73OG3TDWAPGX","created_at":"2026-07-05T11:10:52.426914+00:00"},{"alias_kind":"pith_short_8","alias_value":"IS3K73OG","created_at":"2026-07-05T11:10:52.426914+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.02340","citing_title":"Merging of Bayes and quasi-Bayes empirical Bayes procedures for Poisson compound decisions","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29351","citing_title":"Attention as In-Context Empirical Bayes: A Two-Stage View via Particle Dynamics","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00241","citing_title":"InfoAtlas: A Foundation Model for Zero-Shot Statistical Dependence Estimate","ref_index":239,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03594","citing_title":"Poisson Empirical Bayes via Gamma-Smoothed Nonparametric Maximum Likelihood","ref_index":53,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ","json":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ.json","graph_json":"https://pith.science/api/pith-number/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/graph.json","events_json":"https://pith.science/api/pith-number/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/events.json","paper":"https://pith.science/paper/IS3K73OG"},"agent_actions":{"view_html":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ","download_json":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ.json","view_paper":"https://pith.science/paper/IS3K73OG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.09844&json=true","fetch_graph":"https://pith.science/api/pith-number/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/graph.json","fetch_events":"https://pith.science/api/pith-number/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/action/storage_attestation","attest_author":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/action/author_attestation","sign_citation":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/action/citation_signature","submit_replication":"https://pith.science/pith/IS3K73OG3TDWAPGXZ4T7ZGC4HJ/action/replication_record"}},"created_at":"2026-07-05T11:10:52.426914+00:00","updated_at":"2026-07-05T11:10:52.426914+00:00"}