{"id":"W4404821142","doi":"10.1007/978-3-031-72848-8_14","title":"SPARO: Selective Attention for Robust and Compositional Transformer Encodings for Vision","year":2024,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"Canadian Institute for International Peace and Security; Université de Montréal","funders":"","keywords":"Computer science; Selective attention; Computer vision; Artificial intelligence; Transformer; Neuroscience; Electrical engineering; Cognition; Psychology; Voltage","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":[0.0004665638,0.001224013,0.0009137279,0.0006235122,0.0003440572,0.00137157,0.002035246,0.001063797,0.01753947],"category_scores_gemma":[0.001113947,0.0005052129,0.0008878798,0.0007070535,0.000590725,0.002736578,0.001983759,0.001429932,0.005107649],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006174036,"about_ca_system_score_gemma":0.0006511363,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002328411,"about_ca_topic_score_gemma":0.004617819,"domain_scores_codex":[0.9997378,0.00003322173,0.0000138885,0.00009063345,0.00007313441,0.00005119784],"domain_scores_gemma":[0.999765,0.00007588529,0.00001225522,0.0000867853,0.00003730532,0.00002279309],"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":[0.0005102004,0.000117252,0.0001794376,0.0003252385,0.00007309137,0.0001435769,0.00009100194,0.01564559,0.0747144,0.05986585,0.0463232,0.8020111],"study_design_scores_gemma":[0.00009830359,0.0001889822,0.0003773638,0.00005810968,0.00008197656,0.0003044039,0.00006556142,0.6678621,0.09506991,0.2021307,0.03371432,0.00004826145],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0121446,0.0006761131,0.9582483,0.0001769177,0.0002840712,0.00008575013,0.0007931952,0.01755106,0.01003996],"genre_scores_gemma":[0.2978962,0.0009776137,0.6689373,0.0005703393,0.0002731197,0.0002383425,0.004056696,0.004246678,0.02280364],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01753947,"threshold_uncertainty_score":0.05867535,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01851481912246042,"score_gpt":0.2735095589935203,"score_spread":0.2549947398710599,"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."}}