{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:M5WAGHRJVSWIRP4RJUSBG23B2H","short_pith_number":"pith:M5WAGHRJ","schema_version":"1.0","canonical_sha256":"676c031e29acac88bf914d24136b61d1eeabb00843ee73282896cf71bd132fac","source":{"kind":"arxiv","id":"2608.02050","version":1},"attestation_state":"computed","paper":{"title":"TextNCA: Neural Cellular Automata for Language Modeling via Hierarchical Local Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ashutosh Kumar, Avinash Anand, Avni Mittal, Dikshant Kukreja, Erik Cambria, Kritarth Prasad, Rajiv Ratn Shah, Sushane Dulloo, Timothy Liu, Zhengkui Wang","submitted_at":"2026-08-03T10:47:05Z","abstract_excerpt":"Can a strictly local, iterated, weight-shared computation primitive support language modelling, and which of those three properties actually drives the model's behaviour? We define \\textsc{TextNCA}, a 1D causal windowed-attention realisation of the Neural Cellular Automaton primitive, and study a hierarchical variant that cascades three stages with windows $w \\in \\{8, 32, 128\\}$ and $T_s$ shared-weight iterations per stage, all on WikiText-103 at roughly 30M parameters and 60k training steps. The model does not match a parameter-matched Transformer at this scale (Hier-TextNCA $60.3$ vs.\\ Trans"},"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":"2608.02050","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-03T10:47:05Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"506bf2fdb1c5da456ff694aef29fad3ea79347808fb55f656d09bcabca0ddd99","abstract_canon_sha256":"71eb061fe49654cf57083b0e577c734124bf398ac18b94cc7ec9f9a9ddc62781"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:10:44.797404Z","signature_b64":"Gvk71pFE2O09wcOHg0BvDlgAhSnV7FwEUO/SY8YSFY8CmCPVnBz0+SEtzKOtOW4GodTpEzF8iD884c0AyzV6CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"676c031e29acac88bf914d24136b61d1eeabb00843ee73282896cf71bd132fac","last_reissued_at":"2026-08-04T02:10:44.795914Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:10:44.795914Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TextNCA: Neural Cellular Automata for Language Modeling via Hierarchical Local Attention","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ashutosh Kumar, Avinash Anand, Avni Mittal, Dikshant Kukreja, Erik Cambria, Kritarth Prasad, Rajiv Ratn Shah, Sushane Dulloo, Timothy Liu, Zhengkui Wang","submitted_at":"2026-08-03T10:47:05Z","abstract_excerpt":"Can a strictly local, iterated, weight-shared computation primitive support language modelling, and which of those three properties actually drives the model's behaviour? We define \\textsc{TextNCA}, a 1D causal windowed-attention realisation of the Neural Cellular Automaton primitive, and study a hierarchical variant that cascades three stages with windows $w \\in \\{8, 32, 128\\}$ and $T_s$ shared-weight iterations per stage, all on WikiText-103 at roughly 30M parameters and 60k training steps. The model does not match a parameter-matched Transformer at this scale (Hier-TextNCA $60.3$ vs.\\ Trans"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.02050","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/2608.02050/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":"2608.02050","created_at":"2026-08-04T02:10:44.797396+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.02050v1","created_at":"2026-08-04T02:10:44.797396+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.02050","created_at":"2026-08-04T02:10:44.797396+00:00"},{"alias_kind":"pith_short_12","alias_value":"M5WAGHRJVSWI","created_at":"2026-08-04T02:10:44.797396+00:00"},{"alias_kind":"pith_short_16","alias_value":"M5WAGHRJVSWIRP4R","created_at":"2026-08-04T02:10:44.797396+00:00"},{"alias_kind":"pith_short_8","alias_value":"M5WAGHRJ","created_at":"2026-08-04T02:10:44.797396+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/M5WAGHRJVSWIRP4RJUSBG23B2H","json":"https://pith.science/pith/M5WAGHRJVSWIRP4RJUSBG23B2H.json","graph_json":"https://pith.science/api/pith-number/M5WAGHRJVSWIRP4RJUSBG23B2H/graph.json","events_json":"https://pith.science/api/pith-number/M5WAGHRJVSWIRP4RJUSBG23B2H/events.json","paper":"https://pith.science/paper/M5WAGHRJ"},"agent_actions":{"view_html":"https://pith.science/pith/M5WAGHRJVSWIRP4RJUSBG23B2H","download_json":"https://pith.science/pith/M5WAGHRJVSWIRP4RJUSBG23B2H.json","view_paper":"https://pith.science/paper/M5WAGHRJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.02050&json=true","fetch_graph":"https://pith.science/api/pith-number/M5WAGHRJVSWIRP4RJUSBG23B2H/graph.json","fetch_events":"https://pith.science/api/pith-number/M5WAGHRJVSWIRP4RJUSBG23B2H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M5WAGHRJVSWIRP4RJUSBG23B2H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M5WAGHRJVSWIRP4RJUSBG23B2H/action/storage_attestation","attest_author":"https://pith.science/pith/M5WAGHRJVSWIRP4RJUSBG23B2H/action/author_attestation","sign_citation":"https://pith.science/pith/M5WAGHRJVSWIRP4RJUSBG23B2H/action/citation_signature","submit_replication":"https://pith.science/pith/M5WAGHRJVSWIRP4RJUSBG23B2H/action/replication_record"}},"created_at":"2026-08-04T02:10:44.797396+00:00","updated_at":"2026-08-04T02:10:44.797396+00:00"}