{"id":"W3205791561","doi":"10.48550/arxiv.2110.07752","title":"Hindsight: Posterior-guided training of retrievers for improved open-ended generation","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Hindsight bias; Computer science; Context (archaeology); Generator (circuit theory); Labrador Retriever; Relevance (law); Information retrieval; Artificial intelligence; Natural language processing; Psychology; Cognitive psychology; Medicine; Power (physics); Geography; Political science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004807603,0.001865325,0.001257787,0.0007232888,0.0006074205,0.001334728,0.002960051,0.002699135,0.00794243],"category_scores_gemma":[0.01788582,0.0009479432,0.0009844261,0.000523699,0.001270741,0.002957501,0.003165159,0.003697547,0.003643642],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007746724,"about_ca_system_score_gemma":0.001156776,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00311079,"about_ca_topic_score_gemma":0.006122253,"domain_scores_codex":[0.9977883,0.001058112,0.0001126028,0.0006335737,0.0002402041,0.0001672333],"domain_scores_gemma":[0.9913884,0.006342395,0.0002955175,0.001096209,0.0006379255,0.0002394685],"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":[0.001612348,0.0006869822,0.004727999,0.0006401617,0.0002362479,0.0005790161,0.001194972,0.399337,0.02708547,0.01046273,0.02245224,0.5309849],"study_design_scores_gemma":[0.000138235,0.0001854036,0.0004039559,0.00003436262,0.00003465096,0.0001247669,0.00006234484,0.9789506,0.009783867,0.007756692,0.002500108,0.00002500074],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06471445,0.0009137774,0.9072602,0.0007392059,0.0001713376,0.0002228904,0.0004540702,0.02094783,0.00457628],"genre_scores_gemma":[0.6289619,0.0002259807,0.3523529,0.001336061,0.0001970435,0.0005455842,0.003187835,0.002493139,0.01069961],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00794243,"threshold_uncertainty_score":0.02657008,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.234326412287494,"score_gpt":0.2359855764146311,"score_spread":0.001659164127137125,"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."}}