{"id":"W4404781191","doi":"10.18653/v1/2024.findings-emnlp.677","title":"A Decoding Algorithm for Length-Control Summarization Based on Directed Acyclic Transformers","year":2024,"lang":"en","type":"article","venue":"","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Alliance de recherche numérique du Canada; Alberta Innovates; DeepMind; National Natural Science Foundation of China; Canadian Institute for Advanced Research","keywords":"Automatic summarization; Decoding methods; Computer science; Algorithm; List decoding; Directed acyclic graph; Transformer; Artificial intelligence; Concatenated error correction code; Block code; Voltage; Engineering; Electrical engineering","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002345508,0.000149979,0.0001384644,0.0001874048,0.0001577511,0.0003490845,0.0003454216,0.00006631789,0.0000332637],"category_scores_gemma":[0.00002378698,0.0001152963,0.0001046447,0.0003790128,0.00001287524,0.0006559237,0.00002395011,0.00009365405,0.00002352296],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004878509,"about_ca_system_score_gemma":0.00007422147,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002215724,"about_ca_topic_score_gemma":0.0000048979,"domain_scores_codex":[0.9988248,0.00003282927,0.0001962786,0.000453293,0.0002249633,0.0002678685],"domain_scores_gemma":[0.9992077,0.0003834375,0.00002290742,0.0002549879,0.00004632287,0.00008469573],"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":[0.000007980022,0.00003355525,0.00000674765,0.00001824968,0.00001379515,0.000004271633,0.00004753041,0.0002627746,0.0003723353,0.004321863,0.001955917,0.992955],"study_design_scores_gemma":[0.0006504519,0.0001101507,0.00006209701,0.00006779411,0.00001144048,0.000001449726,0.000004518111,0.9752036,0.001507397,0.0004999031,0.02171884,0.0001623199],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0000546693,0.00008529871,0.9954054,0.0009495869,0.0007036873,0.0003959735,0.00005098667,0.0008881759,0.001466237],"genre_scores_gemma":[0.1869704,0.00003371914,0.8103593,0.00130843,0.0002766105,0.00019375,0.0001579139,0.00003920647,0.0006606849],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9927927,"threshold_uncertainty_score":0.4701645,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01083618685990792,"score_gpt":0.2540388349982392,"score_spread":0.2432026481383313,"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."}}