{"id":"W4416668723","doi":"10.48550/arxiv.2511.17367","title":"R2PS: Worst-Case Robust Real-Time Pursuit Strategies under Partial Observability","year":2025,"lang":"","type":"preprint","venue":"ArXiv.org","topic":"Guidance and Control Systems","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Natural Science Foundation of China; Beijing Nova Program; Institute for Catastrophic Loss Reduction","keywords":"Observability; Reinforcement learning; Pursuer; Generalization; Perfect information; Pursuit-evasion; Complete information; Graph; Markov decision process","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","research_integrity","insufficient_payload"],"consensus_categories":["metaepi_narrow","insufficient_payload"],"category_scores_codex":[0.001603479,0.001705069,0.002271493,0.0002420413,0.0004914554,0.0008210379,0.001559673,0.001625043,0.001284164],"category_scores_gemma":[0.0001908487,0.001872652,0.001036777,0.0006683125,0.0002948984,0.000771287,0.001072224,0.002057627,0.001630563],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008790808,"about_ca_system_score_gemma":0.001071281,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005916302,"about_ca_topic_score_gemma":0.001244933,"domain_scores_codex":[0.9920187,0.0006050231,0.002536657,0.002239678,0.0007659655,0.001833971],"domain_scores_gemma":[0.9945341,0.0004314422,0.0005463358,0.003511934,0.0004479552,0.0005282799],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"observational","study_design_scores_codex":[0.0002481081,0.0004143143,0.09218541,0.003344276,0.001593601,0.001282218,0.001508354,0.8871963,0.006497144,0.002241679,0.001583672,0.00190492],"study_design_scores_gemma":[0.009685751,0.0007345261,0.518723,0.008973161,0.00475999,0.0007992752,0.01717403,0.3955644,0.002799821,0.007030131,0.02163856,0.01211739],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9599934,0.002184275,0.006620881,0.0007008259,0.005501199,0.002190584,0.0002342746,0.001087673,0.02148691],"genre_scores_gemma":[0.9903469,0.0007664168,0.0002903961,0.0001105852,0.00195166,0.0007228801,0.0001144411,0.0001561442,0.005540543],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4916319,"threshold_uncertainty_score":0.999671,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05752719098256209,"score_gpt":0.2645500156708824,"score_spread":0.2070228246883203,"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."}}