{"id":"W4408663284","doi":"10.2139/ssrn.5187495","title":"Enhancing Few-Shot Semantic Segmentation in Remote Sensing Through Magnitude-Based Pruning","year":2025,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Advanced Image and Video Retrieval Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Pruning; Shot (pellet); Segmentation; Magnitude (astronomy); Computer science; Artificial intelligence; Computer vision; Pattern recognition (psychology); Remote sensing; Natural language processing; Geography; Physics; Materials science; Astrophysics","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":[0.00235595,0.000425597,0.0005035055,0.00050633,0.0002609289,0.0003923961,0.00104768,0.0002615741,0.000002676315],"category_scores_gemma":[0.0002064651,0.0004424317,0.0002113172,0.0006966291,0.00003932327,0.000717971,0.0006319946,0.005485609,0.000004639767],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002719703,"about_ca_system_score_gemma":0.00524299,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003111908,"about_ca_topic_score_gemma":0.0009763289,"domain_scores_codex":[0.9955172,0.0003171779,0.0007814445,0.0007264065,0.000513475,0.002144291],"domain_scores_gemma":[0.9983328,0.0001637315,0.0005805066,0.000624783,0.0002384292,0.00005976208],"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":[0.00009292593,0.0000948635,0.0001551975,0.0004438428,0.0002008102,0.0002087726,0.001611845,0.005555451,0.01689071,0.01675421,0.00003238645,0.957959],"study_design_scores_gemma":[0.0008512706,0.0002367031,0.00004880192,0.002579742,0.00005975592,0.0002619329,0.0003968703,0.06193763,0.1299304,0.8027443,0.0002134455,0.00073918],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004685519,0.002891377,0.9899389,0.000771451,0.0004433411,0.0004512629,0.000001257103,0.0002550416,0.0005619227],"genre_scores_gemma":[0.3608672,0.006393695,0.631111,0.0007198381,0.0002612635,0.000003291099,0.00001574371,0.0000442176,0.0005837515],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9572198,"threshold_uncertainty_score":0.9998028,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02040308396928266,"score_gpt":0.3216968579231997,"score_spread":0.3012937739539171,"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."}}