{"id":"W4417283734","doi":"10.1145/3748636.3762753","title":"SpatialGPT: Zero-Shot Vision-and-Language Navigation via Spatial CoT over Structured Spatial Memory","year":2025,"lang":"","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"Natural Sciences and Engineering Research Council of Canada; Alberta Innovates","keywords":"Leverage (statistics); Landmark; Spatial intelligence; Inference; Spatial contextual awareness; Task (project management); Generalization; Natural language; Context (archaeology)","routes":{"ca_aff":true,"ca_fund":true,"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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0007112044,0.0008648592,0.0007953484,0.0005024767,0.0009900751,0.0009230011,0.001602109,0.0005385775,0.001464467],"category_scores_gemma":[0.0002575247,0.0008366862,0.0002587894,0.001182617,0.0004340759,0.0006617192,0.001572545,0.001205953,0.0002037432],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002467901,"about_ca_system_score_gemma":0.0004464261,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.06934907,"about_ca_topic_score_gemma":0.002008047,"domain_scores_codex":[0.9942462,0.0005603433,0.001271517,0.002012357,0.001035349,0.0008741906],"domain_scores_gemma":[0.9960719,0.0005531373,0.0005756814,0.002067906,0.0003206953,0.0004107287],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001243559,0.0002812007,0.005071588,0.0002054107,0.0001605844,0.00002388962,0.003932405,0.002915739,0.03975597,0.01053911,0.0006989448,0.9362908],"study_design_scores_gemma":[0.001749563,0.0001949526,0.2249185,0.000179038,0.0001129414,0.00001984019,0.00004352576,0.760711,0.007230536,0.003680419,0.0004782089,0.0006815619],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2483653,0.0002694795,0.7417316,0.002645393,0.00171919,0.001273392,0.00003782588,0.0003556141,0.003602111],"genre_scores_gemma":[0.9756488,0.00002072621,0.02085992,0.001190179,0.0004412561,0.00009037856,0.0001388077,0.00005526764,0.001554678],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9356092,"threshold_uncertainty_score":0.9994484,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005999108214130407,"score_gpt":0.2956740214560306,"score_spread":0.2896749132419002,"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."}}