{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XBDZJTVPCW3DCKQLZ3OYZVQ7HQ","short_pith_number":"pith:XBDZJTVP","schema_version":"1.0","canonical_sha256":"b84794ceaf15b6312a0bcedd8cd61f3c32c128a0426058d6f879b6da78e8411a","source":{"kind":"arxiv","id":"2606.29171","version":1},"attestation_state":"computed","paper":{"title":"Symbolic Mechanistic Data Attribution: Tracing Training Influence to Learned Behavioral Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Darian Lee, Magy Seif El-Nasr, Reza Habibi","submitted_at":"2026-06-28T03:32:06Z","abstract_excerpt":"While existing data attribution methods can identify which training examples build specific mechanistic circuits, they cannot explain how training data shapes the high-level behavioral decisions a model learns to make. To bridge this gap, we introduce Symbolic Mechanistic Data Attribution (SMDA), a framework that attributes training pairs to the interpretable symbolic policies governing model behavior. SMDA fits a closed-form Ridge regression over sparse autoencoder (SAE) features to model a target behavior, then analytically decomposes how each supervised fine-tuning example shifts that polic"},"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":"2606.29171","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-28T03:32:06Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"6b618be62f95a44d53124cd2c8cf7f50bde0b0edaedf899a98039ddcbea0f8e7","abstract_canon_sha256":"39cc1c64afc3f3b7672c3efd81a64d11fead2ab21326bb7b5676f6bc058796dc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:17:55.523855Z","signature_b64":"rCEj2k3TCCcnR9YUeG4NYWXLlp8ww/JsxOU9mybBFY/lcKU5un6+OiHGrNUgQzF5C5jS3CrMMobvcFneAtYmBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b84794ceaf15b6312a0bcedd8cd61f3c32c128a0426058d6f879b6da78e8411a","last_reissued_at":"2026-06-30T01:17:55.523303Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:17:55.523303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Symbolic Mechanistic Data Attribution: Tracing Training Influence to Learned Behavioral Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Darian Lee, Magy Seif El-Nasr, Reza Habibi","submitted_at":"2026-06-28T03:32:06Z","abstract_excerpt":"While existing data attribution methods can identify which training examples build specific mechanistic circuits, they cannot explain how training data shapes the high-level behavioral decisions a model learns to make. To bridge this gap, we introduce Symbolic Mechanistic Data Attribution (SMDA), a framework that attributes training pairs to the interpretable symbolic policies governing model behavior. SMDA fits a closed-form Ridge regression over sparse autoencoder (SAE) features to model a target behavior, then analytically decomposes how each supervised fine-tuning example shifts that polic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29171","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/2606.29171/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":"2606.29171","created_at":"2026-06-30T01:17:55.523389+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.29171v1","created_at":"2026-06-30T01:17:55.523389+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29171","created_at":"2026-06-30T01:17:55.523389+00:00"},{"alias_kind":"pith_short_12","alias_value":"XBDZJTVPCW3D","created_at":"2026-06-30T01:17:55.523389+00:00"},{"alias_kind":"pith_short_16","alias_value":"XBDZJTVPCW3DCKQL","created_at":"2026-06-30T01:17:55.523389+00:00"},{"alias_kind":"pith_short_8","alias_value":"XBDZJTVP","created_at":"2026-06-30T01:17:55.523389+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/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ","json":"https://pith.science/pith/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ.json","graph_json":"https://pith.science/api/pith-number/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/graph.json","events_json":"https://pith.science/api/pith-number/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/events.json","paper":"https://pith.science/paper/XBDZJTVP"},"agent_actions":{"view_html":"https://pith.science/pith/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ","download_json":"https://pith.science/pith/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ.json","view_paper":"https://pith.science/paper/XBDZJTVP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.29171&json=true","fetch_graph":"https://pith.science/api/pith-number/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/graph.json","fetch_events":"https://pith.science/api/pith-number/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/action/storage_attestation","attest_author":"https://pith.science/pith/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/action/author_attestation","sign_citation":"https://pith.science/pith/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/action/citation_signature","submit_replication":"https://pith.science/pith/XBDZJTVPCW3DCKQLZ3OYZVQ7HQ/action/replication_record"}},"created_at":"2026-06-30T01:17:55.523389+00:00","updated_at":"2026-06-30T01:17:55.523389+00:00"}