{"id":"W7115676588","doi":"10.5267/j.dsl.2025.10.002","title":"Dynamic group fusion transformer for financial time series prediction: An ablation study","year":2025,"lang":"en","type":"article","venue":"Decision Science Letters","topic":"Stock Market Forecasting Methods","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Fusion; Binary number; Architecture; Transformer; Feature (linguistics); Time series; Feature extraction; Artificial neural network","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","sts"],"consensus_categories":[],"category_scores_codex":[0.02144296,0.0002744578,0.0004432486,0.001720932,0.001566986,0.0008908164,0.002025067,0.0001071666,0.00018006],"category_scores_gemma":[0.01365476,0.0002108794,0.000175737,0.005555194,0.0007433469,0.002886169,0.0001681896,0.0001987934,0.00007059159],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002141944,"about_ca_system_score_gemma":0.0002383334,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009350736,"about_ca_topic_score_gemma":0.00005639406,"domain_scores_codex":[0.9931911,0.0003702139,0.001176042,0.001514884,0.003160631,0.0005870839],"domain_scores_gemma":[0.9947545,0.003042566,0.0002639014,0.001168342,0.0005858912,0.0001848187],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0009178911,0.0002512872,0.01134965,0.00000385034,0.00000709516,0.00000566558,0.001326213,0.001116969,0.07914259,0.0004929702,0.008708112,0.8966777],"study_design_scores_gemma":[0.002314373,0.001301489,0.8141687,0.00008387263,0.00005118326,0.00002094449,0.001125856,0.104775,0.001107185,0.05588952,0.01864939,0.0005124855],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5390044,0.000006757155,0.45627,0.001542342,0.001681295,0.0009017144,0.00002086847,0.00007824652,0.0004942842],"genre_scores_gemma":[0.8766928,0.000001945247,0.1201967,0.001746429,0.000113446,0.0001595962,0.00000925143,0.00001940142,0.00106051],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8961652,"threshold_uncertainty_score":0.9997329,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04222262798693358,"score_gpt":0.3941564552567649,"score_spread":0.3519338272698313,"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."}}