{"id":"W7084061681","doi":"10.24963/ijcai.2024/1271","title":"SandboxSocial: A Sandbox for Social Media Using Multimodal AI Agents","year":2024,"lang":"en","type":"article","venue":"","topic":"Canadian Policy and Governance","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Impact; Université de Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Social media; Sandbox (software development); Bridge (graph theory); Key (lock); Grounded theory; Upload; Social dynamics","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":[0.001346499,0.0009637783,0.0005352857,0.0006952351,0.001216438,0.001749088,0.002527041,0.001746325,0.01513186],"category_scores_gemma":[0.006243537,0.0005691739,0.0009116021,0.0003835007,0.001108395,0.002618589,0.003303972,0.001509327,0.001771162],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001149396,"about_ca_system_score_gemma":0.001973587,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0111224,"about_ca_topic_score_gemma":0.01293007,"domain_scores_codex":[0.999435,0.0002445137,0.00004119548,0.00009508929,0.0001211271,0.00006303008],"domain_scores_gemma":[0.9976103,0.001191953,0.0001124928,0.00053803,0.0002209469,0.0003261798],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001109145,0.0009698938,0.01419286,0.0007908448,0.000376487,0.000855512,0.001898901,0.7174639,0.01063208,0.1284408,0.04606026,0.07720929],"study_design_scores_gemma":[0.0001426821,0.00009769687,0.000499326,0.00004321007,0.00002932237,0.00006233702,0.0001207445,0.9425285,0.002452082,0.02223836,0.03174642,0.00003928278],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1764066,0.0006018397,0.6751828,0.002447405,0.00073669,0.002152892,0.006044333,0.07413998,0.0622875],"genre_scores_gemma":[0.708932,0.0005073942,0.2627196,0.0006141623,0.00007349747,0.002183171,0.003925751,0.003914172,0.01713016],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01513186,"threshold_uncertainty_score":0.05062109,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08376147502815687,"score_gpt":0.3979324089913455,"score_spread":0.3141709339631886,"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."}}