{"id":"W7111103067","doi":"10.5281/zenodo.17850893","title":"The Collapse Index (CI) CrackTest: Morphology-Aligned Perturbation Testing Reveals Systematic Collapse Inheritance in Frontier Language Models","year":2025,"lang":"en","type":"preprint","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Glycemic Index Laboratories","funders":"","keywords":"Inference; Bounded function; Robustness (evolution); Perturbation (astronomy); Observable; Progressive collapse","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003808691,0.0009876767,0.000548342,0.001243719,0.0005594477,0.001939713,0.00141899,0.0009384088,0.003050357],"category_scores_gemma":[0.02356815,0.0003849534,0.0008964262,0.0004788481,0.00162317,0.002440086,0.002634785,0.001949671,0.0008136053],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000963797,"about_ca_system_score_gemma":0.001592976,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003008147,"about_ca_topic_score_gemma":0.00341268,"domain_scores_codex":[0.9964566,0.001506237,0.0001682827,0.0006105799,0.001049351,0.0002089707],"domain_scores_gemma":[0.987964,0.007438964,0.0008017292,0.002489196,0.001020579,0.0002854926],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001553377,0.0003389368,0.02922946,0.0004898966,0.0003622041,0.0009122522,0.001254526,0.5261137,0.07907905,0.07259831,0.01443271,0.2736356],"study_design_scores_gemma":[0.00002678456,0.0001890407,0.00261921,0.0000299608,0.00002750805,0.0001507614,0.0001288965,0.9378646,0.02002407,0.03717046,0.001723721,0.00004499639],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2091741,0.0001812923,0.7738673,0.0003773763,0.00009127632,0.000145175,0.001505044,0.01061625,0.004042226],"genre_scores_gemma":[0.8631113,0.00007390738,0.131402,0.0001472711,0.00003559034,0.0001472259,0.002491779,0.001605165,0.0009858308],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003808691,"threshold_uncertainty_score":0.02014256,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03268787693774515,"score_gpt":0.2676161464680126,"score_spread":0.2349282695302674,"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."}}