{"id":"W4411586529","doi":"10.1017/dap.2026.10064","title":"The Attribution Crisis in LLM Search Results: Estimating Ecosystem Exploitation","year":2025,"lang":"en","type":"preprint","venue":"Data & Policy","topic":"Research Data Management Practices","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Alfred P. Sloan Foundation; Patrick J. McGovern Foundation","keywords":"Attribution; Ecosystem; Environmental science; Ecology; Psychology; Biology; Social psychology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.02528711,0.0007430872,0.0008509608,0.007990285,0.0006662917,0.00416925,0.001349955,0.001676379,0.001434319],"category_scores_gemma":[0.1698516,0.0005044119,0.0008023404,0.007843793,0.001465672,0.008380172,0.0029026,0.001843635,0.001094947],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001981713,"about_ca_system_score_gemma":0.0008339186,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01118562,"about_ca_topic_score_gemma":0.004995769,"domain_scores_codex":[0.9873076,0.006110116,0.001566394,0.001641337,0.002698289,0.0006761392],"domain_scores_gemma":[0.8384038,0.1246418,0.01580293,0.0120893,0.006764119,0.002298122],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0007233206,0.0002142159,0.9063883,0.0002409387,0.0002762779,0.0001859594,0.001303893,0.05139333,0.0007532069,0.003674662,0.004362735,0.03048303],"study_design_scores_gemma":[0.00003803669,0.0001869043,0.2725495,0.0001082246,0.000107856,0.0005130225,0.001514168,0.7082943,0.002668488,0.01091588,0.002993069,0.0001104241],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9834731,0.0007461698,0.008065173,0.0009488928,0.00002940595,0.00006453846,0.003141834,0.0006715368,0.002859303],"genre_scores_gemma":[0.9961846,0.00006516871,0.001944486,0.00004554298,0.00002171697,0.00002311656,0.00149258,0.00004691894,0.0001757544],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9747129,"threshold_uncertainty_score":0.1337326,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2467715250817659,"score_gpt":0.466083943034606,"score_spread":0.2193124179528401,"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."}}