{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EJUUNUDY72UMNRSC2U7IGHFL6J","short_pith_number":"pith:EJUUNUDY","schema_version":"1.0","canonical_sha256":"226946d078fea8c6c642d53e831cabf249b2d371c5730acb7165299e0c1d8fa6","source":{"kind":"arxiv","id":"2506.16550","version":3},"attestation_state":"computed","paper":{"title":"A Free Probabilistic Framework for Analyzing the Transformer-based Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Swagatam Das","submitted_at":"2025-06-19T19:13:02Z","abstract_excerpt":"We present a formal operator-theoretic framework for analyzing Transformer-based language models using free probability theory. By modeling token embeddings and attention mechanisms as self-adjoint operators in a tracial \\( W^* \\)-probability space, we reinterpret attention as non-commutative convolution and describe representation propagation via free additive convolution. This leads to a spectral dynamic system interpretation of deep Transformers. We derive entropy-based generalization bounds under freeness assumptions and provide insight into positional encoding, spectral evolution, and rep"},"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":"2506.16550","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-19T19:13:02Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"c783457aaca3269eba45b39e5ac4abe2707e8eed40722c108825438e1bad4b1d","abstract_canon_sha256":"5f224f38009a2ef5670daa38ee53df7ed620dec31165e6457a32e232f6e3325f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:48.528113Z","signature_b64":"MA69gI6mL6OO0mz+xOB7P9IF+PrJwg+osN/8o32IHeXrUiNpDmvJp6Rosbd3zoK0HqLoXmp2bdIUKkbukx/4DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"226946d078fea8c6c642d53e831cabf249b2d371c5730acb7165299e0c1d8fa6","last_reissued_at":"2026-07-05T11:54:48.527696Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:48.527696Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Free Probabilistic Framework for Analyzing the Transformer-based Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Swagatam Das","submitted_at":"2025-06-19T19:13:02Z","abstract_excerpt":"We present a formal operator-theoretic framework for analyzing Transformer-based language models using free probability theory. By modeling token embeddings and attention mechanisms as self-adjoint operators in a tracial \\( W^* \\)-probability space, we reinterpret attention as non-commutative convolution and describe representation propagation via free additive convolution. This leads to a spectral dynamic system interpretation of deep Transformers. We derive entropy-based generalization bounds under freeness assumptions and provide insight into positional encoding, spectral evolution, and rep"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.16550","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/2506.16550/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":"2506.16550","created_at":"2026-07-05T11:54:48.527752+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.16550v3","created_at":"2026-07-05T11:54:48.527752+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.16550","created_at":"2026-07-05T11:54:48.527752+00:00"},{"alias_kind":"pith_short_12","alias_value":"EJUUNUDY72UM","created_at":"2026-07-05T11:54:48.527752+00:00"},{"alias_kind":"pith_short_16","alias_value":"EJUUNUDY72UMNRSC","created_at":"2026-07-05T11:54:48.527752+00:00"},{"alias_kind":"pith_short_8","alias_value":"EJUUNUDY","created_at":"2026-07-05T11:54:48.527752+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.06776","citing_title":"Zero-Direction Probing: A Linear-Algebraic Framework for Deep Analysis of Large-Language-Model Drift","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J","json":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J.json","graph_json":"https://pith.science/api/pith-number/EJUUNUDY72UMNRSC2U7IGHFL6J/graph.json","events_json":"https://pith.science/api/pith-number/EJUUNUDY72UMNRSC2U7IGHFL6J/events.json","paper":"https://pith.science/paper/EJUUNUDY"},"agent_actions":{"view_html":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J","download_json":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J.json","view_paper":"https://pith.science/paper/EJUUNUDY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.16550&json=true","fetch_graph":"https://pith.science/api/pith-number/EJUUNUDY72UMNRSC2U7IGHFL6J/graph.json","fetch_events":"https://pith.science/api/pith-number/EJUUNUDY72UMNRSC2U7IGHFL6J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J/action/storage_attestation","attest_author":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J/action/author_attestation","sign_citation":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J/action/citation_signature","submit_replication":"https://pith.science/pith/EJUUNUDY72UMNRSC2U7IGHFL6J/action/replication_record"}},"created_at":"2026-07-05T11:54:48.527752+00:00","updated_at":"2026-07-05T11:54:48.527752+00:00"}