{"id":"W7092316905","doi":"","title":"Spectral Analysis of Molecular Features: When Richer Features Do Not Guarantee Better Generalization","year":2025,"lang":"","type":"article","venue":"ArXiv.org","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Alliance de recherche numérique du Canada","keywords":"Kernel (algebra); Property (philosophy); Pattern recognition (psychology); Generalization; Kernel method; Eigenvalues and eigenvectors; Robustness (evolution); Correlation; Representation (politics)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001939373,0.0007288713,0.001369692,0.001217196,0.0004263163,0.0005987812,0.001888207,0.0004686152,0.004404942],"category_scores_gemma":[0.0009061886,0.0006760479,0.0005915543,0.003501698,0.0006908783,0.0005178659,0.0006059263,0.0005016546,0.0001731837],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001826326,"about_ca_system_score_gemma":0.0002458746,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001117035,"about_ca_topic_score_gemma":0.0001518792,"domain_scores_codex":[0.9936401,0.001128813,0.001329957,0.001692514,0.001145603,0.001063037],"domain_scores_gemma":[0.9965193,0.0002311536,0.0008926157,0.001785925,0.0004316824,0.0001392733],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"observational","study_design_scores_codex":[0.0001432316,0.0001330879,0.1873418,0.0001570905,0.0005641691,0.00003163352,0.001200509,0.02932167,0.7770832,0.002256384,0.001569428,0.0001978549],"study_design_scores_gemma":[0.0004653939,0.00008951424,0.5147168,0.0001522067,0.001860677,0.000003743838,0.00006938581,0.003086095,0.4777345,0.0007809824,0.0005279906,0.0005127043],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.982295,0.00133538,0.007902508,0.00379822,0.002255544,0.0005243772,0.0001019666,0.0001132879,0.001673763],"genre_scores_gemma":[0.9844441,0.0001484047,0.007461393,0.003415862,0.0002430546,0.0000289935,0.00006071611,0.00005105175,0.004146434],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.327375,"threshold_uncertainty_score":0.9995691,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01178208830966216,"score_gpt":0.2765770003173438,"score_spread":0.2647949120076816,"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."}}