{"id":"W4408867802","doi":"10.2139/ssrn.5193979","title":"V-Words, Lyndon Words and Galois Words","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McMaster University","funders":"","keywords":"Mathematics; Computer science; Natural language processing; Linguistics; Arithmetic; Philosophy","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.002030217,0.0004926955,0.0005487273,0.0004018296,0.00043283,0.0008172423,0.002548169,0.0003730991,0.00002390646],"category_scores_gemma":[0.0000592838,0.0004283652,0.0002198928,0.0003723789,0.00006834534,0.0005324568,0.00366526,0.006429554,0.00001272167],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009066509,"about_ca_system_score_gemma":0.00518102,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002527799,"about_ca_topic_score_gemma":0.0003030616,"domain_scores_codex":[0.9951282,0.0002137688,0.0005746005,0.0009164086,0.0006305131,0.00253648],"domain_scores_gemma":[0.9980304,0.0001045419,0.0004208927,0.001114216,0.0001108798,0.0002191283],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00005085571,0.0001325654,0.0005424655,0.00008542616,0.0003591954,0.00004143756,0.0008718283,0.0002200445,0.00002927822,0.1834945,0.002946805,0.8112256],"study_design_scores_gemma":[0.001247375,0.000303491,0.0005160746,0.0008850973,0.0001059163,0.0009356879,0.00042047,0.05445384,0.00006661253,0.9122403,0.02776114,0.001064054],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005231095,0.02236051,0.9656951,0.001771467,0.002244956,0.0002644429,0.00002388825,0.0001706241,0.00223785],"genre_scores_gemma":[0.7720717,0.1377788,0.05988674,0.001208266,0.004060932,0.00009633182,0.0001217302,0.000120848,0.02465466],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9058084,"threshold_uncertainty_score":0.9998168,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008411261213417329,"score_gpt":0.2528590531951012,"score_spread":0.2444477919816838,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}