{"id":"W7126443349","doi":"10.18653/v1/2024.clpsych-1.15","title":"Overview of the CLPsych 2024 Shared Task: Leveraging Large Language Models to Identify Evidence of Suicidality Risk in Online Posts","year":2024,"lang":"","type":"article","venue":"","topic":"Mental Health via Writing","field":"Psychology","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Engineering and Physical Sciences Research Council; Atomic Energy of Canada Limited; Bar-Ilan University; UK Research and Innovation; National Science Foundation","keywords":"Computational linguistics; The Internet; Poison control; On Language; Computational model; Identification (biology)","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.005431443,0.0004674108,0.000951028,0.0004378151,0.0001222355,0.00007982085,0.001123429,0.0003129298,0.007426216],"category_scores_gemma":[0.0003491682,0.0003940364,0.0003936242,0.002373494,0.0001030913,0.00044305,0.0009690078,0.001260214,0.0002012512],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003813615,"about_ca_system_score_gemma":0.000354965,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02106006,"about_ca_topic_score_gemma":0.009538938,"domain_scores_codex":[0.9922861,0.001892939,0.002589924,0.001231717,0.000901718,0.001097617],"domain_scores_gemma":[0.9962085,0.0012312,0.0005244961,0.001632807,0.0001493024,0.0002537044],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"observational","study_design_scores_codex":[0.001986984,0.007271349,0.2831134,0.08677236,0.001217195,0.0006332493,0.3834579,0.001572054,0.03640152,0.03539466,0.01793936,0.1442399],"study_design_scores_gemma":[0.002916983,0.0006511017,0.7923045,0.07565947,0.000457939,0.0000811682,0.05177007,0.06580171,0.003586195,0.004893569,0.0006255397,0.00125174],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.889916,0.09676765,0.0006451529,0.002630433,0.002552252,0.001913105,0.002509325,0.00004844413,0.003017647],"genre_scores_gemma":[0.9932154,0.001707298,0.0005384232,0.001658372,0.0001658261,0.00004386244,0.00001884499,0.00007332997,0.002578602],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5091911,"threshold_uncertainty_score":0.9998512,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1862968720971332,"score_gpt":0.4930726363834393,"score_spread":0.306775764286306,"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."}}