{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:LCY6BKW3OELAAQIT5XGU2CH7BI","short_pith_number":"pith:LCY6BKW3","schema_version":"1.0","canonical_sha256":"58b1e0aadb7116004113edcd4d08ff0a1a313d3f9454ffa2199931544ce493aa","source":{"kind":"arxiv","id":"2505.00033","version":1},"attestation_state":"computed","paper":{"title":"From Attention to Atoms: Spectral Dictionary Learning for Fast, Interpretable Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andrew Kiruluta","submitted_at":"2025-04-29T13:24:42Z","abstract_excerpt":"We propose a novel spectral generative modeling framework for natural language processing that jointly learns a global time varying Fourier dictionary and per token mixing coefficients, replacing the ubiquitous self attention mechanism in transformer architectures. By enforcing reconstruction losses in both the time domain (embedding reconstruction) and the frequency domain (via Short Time Fourier Transform magnitude matching) alongside a standard language modeling objective, and fitting a Gaussian Mixture Model (GMM) prior over the learned mixing vectors, our approach achieves competitive per"},"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":"2505.00033","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-29T13:24:42Z","cross_cats_sorted":[],"title_canon_sha256":"dfe831950036a8a6a58b64d4a8dc7c7fcd1667ee9c998782904a50c014d968a8","abstract_canon_sha256":"fbe0d538806ae48d148ae01003fcc284e4388c967acfefd60207a41e1a340cb0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:56:40.818085Z","signature_b64":"43UMAf+thDEcTuyJ3XRquj++scVdTy1B98yUKbo4QN8Dx9S3Tw6qgKq2ZxwA/Xz2bUBxpaWBgX3s0RJXse13Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58b1e0aadb7116004113edcd4d08ff0a1a313d3f9454ffa2199931544ce493aa","last_reissued_at":"2026-07-05T10:56:40.814982Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:56:40.814982Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"From Attention to Atoms: Spectral Dictionary Learning for Fast, Interpretable Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andrew Kiruluta","submitted_at":"2025-04-29T13:24:42Z","abstract_excerpt":"We propose a novel spectral generative modeling framework for natural language processing that jointly learns a global time varying Fourier dictionary and per token mixing coefficients, replacing the ubiquitous self attention mechanism in transformer architectures. By enforcing reconstruction losses in both the time domain (embedding reconstruction) and the frequency domain (via Short Time Fourier Transform magnitude matching) alongside a standard language modeling objective, and fitting a Gaussian Mixture Model (GMM) prior over the learned mixing vectors, our approach achieves competitive per"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.00033","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/2505.00033/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":"2505.00033","created_at":"2026-07-05T10:56:40.817489+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.00033v1","created_at":"2026-07-05T10:56:40.817489+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.00033","created_at":"2026-07-05T10:56:40.817489+00:00"},{"alias_kind":"pith_short_12","alias_value":"LCY6BKW3OELA","created_at":"2026-07-05T10:56:40.817489+00:00"},{"alias_kind":"pith_short_16","alias_value":"LCY6BKW3OELAAQIT","created_at":"2026-07-05T10:56:40.817489+00:00"},{"alias_kind":"pith_short_8","alias_value":"LCY6BKW3","created_at":"2026-07-05T10:56:40.817489+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.19514","citing_title":"Wavelet Logic Machines: Learning and Reasoning in the Spectral Domain Without Neural Networks","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI","json":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI.json","graph_json":"https://pith.science/api/pith-number/LCY6BKW3OELAAQIT5XGU2CH7BI/graph.json","events_json":"https://pith.science/api/pith-number/LCY6BKW3OELAAQIT5XGU2CH7BI/events.json","paper":"https://pith.science/paper/LCY6BKW3"},"agent_actions":{"view_html":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI","download_json":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI.json","view_paper":"https://pith.science/paper/LCY6BKW3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.00033&json=true","fetch_graph":"https://pith.science/api/pith-number/LCY6BKW3OELAAQIT5XGU2CH7BI/graph.json","fetch_events":"https://pith.science/api/pith-number/LCY6BKW3OELAAQIT5XGU2CH7BI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI/action/storage_attestation","attest_author":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI/action/author_attestation","sign_citation":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI/action/citation_signature","submit_replication":"https://pith.science/pith/LCY6BKW3OELAAQIT5XGU2CH7BI/action/replication_record"}},"created_at":"2026-07-05T10:56:40.817489+00:00","updated_at":"2026-07-05T10:56:40.817489+00:00"}