{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:Q4RR4HSLDSD4KK727UGOD6WFIN","short_pith_number":"pith:Q4RR4HSL","schema_version":"1.0","canonical_sha256":"87231e1e4b1c87c52bfafd0ce1fac5437141f326a1cdbcc4efe9e928ccfefbb0","source":{"kind":"arxiv","id":"2502.10648","version":3},"attestation_state":"computed","paper":{"title":"LLM-Lasso: A Robust Framework for Domain-Informed Feature Selection and Regularization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ash Alizadeh, Erica Zhang, Jose Blanchet, Jurik Mutter, Kangwook Lee, Mert Pilanci, Naomi Sagan, Nick Phillips, Robert Tibshirani, Ryunosuke Goto","submitted_at":"2025-02-15T02:55:22Z","abstract_excerpt":"We introduce LLM-Lasso, a novel framework that leverages large language models (LLMs) to guide feature selection in Lasso $\\ell_1$ regression. Unlike traditional methods that rely solely on numerical data, LLM-Lasso incorporates domain-specific knowledge extracted from natural language, enhanced through a retrieval-augmented generation (RAG) pipeline, to seamlessly integrate data-driven modeling with contextual insights. Specifically, the LLM generates penalty factors for each feature, which are converted into weights for the Lasso penalty using a simple, tunable model. Features identified as "},"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":"2502.10648","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-15T02:55:22Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"142b86bc93dd1ab7a180a7bd3c620dcd06698a91672c39fb892cd93aa62b705d","abstract_canon_sha256":"c86affa51ac0ad17f1ee9f310bcaa7e1efeb26fed142403be1694fa7418b4865"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:52:17.824043Z","signature_b64":"MockUSrbd4XRZ2D1sxinBJsxOEQpUe48Ru8MbrOwXFlFHT0qB71dMoNWjKbbBr0OBEDiRMiiv/8Q1+wrYprHBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87231e1e4b1c87c52bfafd0ce1fac5437141f326a1cdbcc4efe9e928ccfefbb0","last_reissued_at":"2026-07-05T11:52:17.823533Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:52:17.823533Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM-Lasso: A Robust Framework for Domain-Informed Feature Selection and Regularization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ash Alizadeh, Erica Zhang, Jose Blanchet, Jurik Mutter, Kangwook Lee, Mert Pilanci, Naomi Sagan, Nick Phillips, Robert Tibshirani, Ryunosuke Goto","submitted_at":"2025-02-15T02:55:22Z","abstract_excerpt":"We introduce LLM-Lasso, a novel framework that leverages large language models (LLMs) to guide feature selection in Lasso $\\ell_1$ regression. Unlike traditional methods that rely solely on numerical data, LLM-Lasso incorporates domain-specific knowledge extracted from natural language, enhanced through a retrieval-augmented generation (RAG) pipeline, to seamlessly integrate data-driven modeling with contextual insights. Specifically, the LLM generates penalty factors for each feature, which are converted into weights for the Lasso penalty using a simple, tunable model. Features identified as "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.10648","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/2502.10648/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":"2502.10648","created_at":"2026-07-05T11:52:17.823594+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.10648v3","created_at":"2026-07-05T11:52:17.823594+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.10648","created_at":"2026-07-05T11:52:17.823594+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q4RR4HSLDSD4","created_at":"2026-07-05T11:52:17.823594+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q4RR4HSLDSD4KK72","created_at":"2026-07-05T11:52:17.823594+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q4RR4HSL","created_at":"2026-07-05T11:52:17.823594+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.29516","citing_title":"A Mathematical Optimization Approach for Expert-Informed Bayesian Best Subset Selection","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19464","citing_title":"LePREC: Reasoning as Classification over Structured Factors for Assessing Relevance of Legal Issues","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14334","citing_title":"Mamba-SSM with LLM Reasoning for Feature Selection: Faithfulness-Aware Biomarker Discovery","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN","json":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN.json","graph_json":"https://pith.science/api/pith-number/Q4RR4HSLDSD4KK727UGOD6WFIN/graph.json","events_json":"https://pith.science/api/pith-number/Q4RR4HSLDSD4KK727UGOD6WFIN/events.json","paper":"https://pith.science/paper/Q4RR4HSL"},"agent_actions":{"view_html":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN","download_json":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN.json","view_paper":"https://pith.science/paper/Q4RR4HSL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.10648&json=true","fetch_graph":"https://pith.science/api/pith-number/Q4RR4HSLDSD4KK727UGOD6WFIN/graph.json","fetch_events":"https://pith.science/api/pith-number/Q4RR4HSLDSD4KK727UGOD6WFIN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN/action/storage_attestation","attest_author":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN/action/author_attestation","sign_citation":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN/action/citation_signature","submit_replication":"https://pith.science/pith/Q4RR4HSLDSD4KK727UGOD6WFIN/action/replication_record"}},"created_at":"2026-07-05T11:52:17.823594+00:00","updated_at":"2026-07-05T11:52:17.823594+00:00"}