{"id":"W4400104527","doi":"10.48550/arxiv.2406.18043","title":"GenRL: Multimodal-foundation world models for generalization in embodied agents","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Multi-Agent Systems and Negotiation","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Vlaamse regering; Mitacs; Fonds Wetenschappelijk Onderzoek","keywords":"Embodied cognition; Foundation (evidence); Generalist and specialist species; Computer science; Artificial intelligence; Geography; Ecology; Biology; Archaeology; Habitat","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"],"consensus_categories":[],"category_scores_codex":[0.0003556825,0.0002918939,0.000299838,0.0007710065,0.0001087367,0.0002333934,0.0008014533,0.0002264571,0.0000168489],"category_scores_gemma":[0.00001634245,0.0003456839,0.0001866979,0.0007264066,0.00002109685,0.0005382145,0.0008681024,0.0002702303,0.00008035123],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005002688,"about_ca_system_score_gemma":0.0001592481,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006295343,"about_ca_topic_score_gemma":0.001008034,"domain_scores_codex":[0.9979401,0.0001281877,0.0003533204,0.001157252,0.0001177221,0.0003033804],"domain_scores_gemma":[0.9987097,0.00005547598,0.00026633,0.0007191956,0.0001568093,0.00009245517],"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":[0.00001203091,0.00005248002,0.0005540485,0.0001934381,0.00003646101,0.00001983551,0.0004135769,0.7266101,0.00004361182,0.2710354,0.0003635121,0.0006655643],"study_design_scores_gemma":[0.0005311008,0.00001210251,0.0006299503,0.0001522344,0.00003302414,4.461289e-7,0.00001344179,0.8966388,0.00007530052,0.1012222,0.0003798626,0.000311475],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09651995,0.00005546714,0.8988948,0.00009996932,0.00158889,0.001049869,0.00001865144,0.0002234575,0.001548935],"genre_scores_gemma":[0.9908972,0.00007286069,0.004537681,0.00007493996,0.0001331092,0.00001383901,0.0001683539,0.00002860979,0.004073444],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8943772,"threshold_uncertainty_score":0.9998995,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1369722439224506,"score_gpt":0.2332615442925057,"score_spread":0.09628930037005509,"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."}}