{"id":"W4412122873","doi":"10.1073/pnas.2502599122","title":"Asymptotic theory of in-context learning by linear attention","year":2025,"lang":"en","type":"article","venue":"Proceedings of the National Academy of Sciences","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada); Perimeter Institute","funders":"National Science Foundation","keywords":"Learning curve; Computer science; Generalization; Memorization; Context (archaeology); Security token; Task (project management); Scaling; Mathematics; Artificial intelligence; Machine learning; Cognitive psychology; Psychology","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":[],"consensus_categories":[],"category_scores_codex":[0.00247528,0.00006458575,0.00013078,0.0002791306,0.0001215364,0.00002434737,0.001022274,0.00005321774,0.000005810626],"category_scores_gemma":[0.0008740979,0.00004833449,0.00006027365,0.00134442,0.0003892953,0.0005449082,0.0001824468,0.0001906331,0.000001202552],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002524182,"about_ca_system_score_gemma":0.00004179828,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003377611,"about_ca_topic_score_gemma":4.782033e-8,"domain_scores_codex":[0.9986352,0.00002480272,0.0003480673,0.0002048388,0.0006713246,0.0001157434],"domain_scores_gemma":[0.9991204,0.000249886,0.0003969197,0.000008928751,0.0002079203,0.00001598324],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.000009611455,0.00004725649,0.02215332,0.00005728043,0.00001066113,2.933421e-9,0.0006119826,0.000885367,0.1206047,0.8460068,0.0001840057,0.009429038],"study_design_scores_gemma":[0.0009331195,0.0001345181,0.3487162,0.000749528,0.000015319,0.000003857964,0.002741793,0.1426581,0.1862842,0.3159348,0.001589184,0.0002394197],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9621997,0.0004166105,0.003482438,0.004367259,0.00006387514,0.0002204557,0.000001701379,0.00003144714,0.02921646],"genre_scores_gemma":[0.9958404,0.00001485561,0.002981689,0.0002321135,0.000008534908,0.000003386725,5.686453e-8,0.000001492735,0.0009174352],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.530072,"threshold_uncertainty_score":0.1971023,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02662637391215992,"score_gpt":0.2931535029031697,"score_spread":0.2665271289910098,"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."}}