{"id":"W4385571655","doi":"10.18653/v1/2023.findings-acl.46","title":"The Web Can Be Your Oyster for Improving Language Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"Renmin University of China; National Natural Science Foundation of China","keywords":"Computer science; Salient; Language model; Task (project management); Information retrieval; ENCODE; Machine learning; Search engine; Artificial intelligence","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":[0.001727963,0.001643632,0.0007487679,0.001541263,0.000523473,0.001740469,0.001568542,0.001331613,0.009367543],"category_scores_gemma":[0.008435614,0.0006748433,0.0013027,0.001417565,0.0004964772,0.006944722,0.002017867,0.002789912,0.006746029],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005766524,"about_ca_system_score_gemma":0.001026243,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009467509,"about_ca_topic_score_gemma":0.01459961,"domain_scores_codex":[0.999364,0.0002072077,0.00004403804,0.0002137342,0.0001259102,0.00004512228],"domain_scores_gemma":[0.9973164,0.001259806,0.00009942602,0.0008709243,0.0003584168,0.00009495661],"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":[0.0004236213,0.0004442845,0.004127015,0.0004036893,0.0003017228,0.0002320546,0.000197023,0.1442786,0.01136496,0.005974155,0.04507676,0.787176],"study_design_scores_gemma":[0.00005271378,0.0001605486,0.001247932,0.00008345421,0.0000937867,0.00010644,0.00007613739,0.9525538,0.008593861,0.01484242,0.02214251,0.00004628168],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07022582,0.008681216,0.8560348,0.003761984,0.0009565097,0.000211307,0.003851487,0.03998687,0.01628998],"genre_scores_gemma":[0.5304942,0.005000181,0.4293673,0.002049023,0.0005476808,0.0004226792,0.01136312,0.002499581,0.01825622],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009467509,"threshold_uncertainty_score":0.03133762,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05366418091734578,"score_gpt":0.280569888942892,"score_spread":0.2269057080255462,"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."}}