{"id":"W3096316167","doi":"10.48550/arxiv.2011.02944","title":"Learning Efficient Task-Specific Meta-Embeddings with Word Prisms","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Word (group theory); Computer science; Embedding; Inference; Set (abstract data type); Task (project management); Natural language processing; Artificial intelligence; Word embedding; Context (archaeology); Simple (philosophy); Meta learning (computer science); Space (punctuation); Mathematics","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":[0.002109894,0.002344594,0.001167468,0.001356432,0.0003454186,0.001644838,0.002383025,0.001582825,0.002773828],"category_scores_gemma":[0.007671096,0.0009084832,0.001860845,0.002089467,0.000781886,0.007922399,0.003739197,0.003173033,0.0024914],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006146072,"about_ca_system_score_gemma":0.001418789,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001537843,"about_ca_topic_score_gemma":0.003325554,"domain_scores_codex":[0.9986085,0.0004725545,0.0001418483,0.0004464683,0.0002222296,0.0001083503],"domain_scores_gemma":[0.9971952,0.0009877845,0.0002272775,0.001070644,0.0003979532,0.0001210714],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006280782,0.000468005,0.004989441,0.0007928113,0.0004284725,0.0002648789,0.0004495826,0.217773,0.01928721,0.01938953,0.02027367,0.7152554],"study_design_scores_gemma":[0.00007485211,0.0001646143,0.0005048619,0.00004080976,0.00006934682,0.0001263357,0.0001043034,0.9526964,0.008684302,0.03354448,0.003956234,0.00003340861],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0384731,0.0009692939,0.9458401,0.0003812913,0.0001814028,0.0001187437,0.001275913,0.01115729,0.001602953],"genre_scores_gemma":[0.4524687,0.0008231638,0.5301941,0.0003299931,0.0001662922,0.0004558654,0.009049787,0.00128078,0.005231258],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002773828,"threshold_uncertainty_score":0.01115835,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1082125117793126,"score_gpt":0.1844184749974074,"score_spread":0.07620596321809482,"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."}}