{"id":"W4410004015","doi":"10.36227/techrxiv.174612962.26131807/v1","title":"A Survey on Large Language Model based Human-Agent Systems","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Robotics and Automated Systems","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005770251,0.0003760993,0.0005345106,0.0002211648,0.00005701169,0.0001639297,0.000355066,0.0003910667,0.00003817532],"category_scores_gemma":[0.00001574813,0.0003462588,0.0001412772,0.0001021088,0.000005753102,0.00001417788,0.0001482334,0.0004730283,0.00007703632],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001740537,"about_ca_system_score_gemma":0.00006383341,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009585478,"about_ca_topic_score_gemma":0.0002852089,"domain_scores_codex":[0.9984463,0.0001035748,0.0004607208,0.0003673157,0.0002712591,0.0003507921],"domain_scores_gemma":[0.9988933,0.00006585514,0.00005968233,0.0008408896,0.00005819184,0.0000820508],"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":[0.000001254681,0.00003779435,0.0000797336,0.001034241,0.00008729291,0.0000107124,0.00008890898,0.9809784,0.0001323466,0.0009060254,0.01662048,0.00002275242],"study_design_scores_gemma":[0.000228805,0.000007694266,0.0003618639,0.0006701826,0.00001776174,1.443383e-7,0.00002974644,0.9979398,0.0001806698,0.000008756895,0.0002104444,0.0003441941],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0580065,0.002596897,0.7438013,0.00003092955,0.006985083,0.002239594,0.004874601,0.005116591,0.1763485],"genre_scores_gemma":[0.9880267,0.00001079401,0.0002085107,0.00004999464,0.00008100809,0.00008668227,0.0007975044,0.00005801077,0.01068084],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9300202,"threshold_uncertainty_score":0.999899,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03213525556568443,"score_gpt":0.2812408818956744,"score_spread":0.2491056263299899,"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."}}