{"meta":{"query_hash":"add399e4d669","filters":{"venue":"Linguistic Data Consortium Catalog"},"cohort_total":2,"direct_labels_cover":0,"predictions_cover":2,"exported":2,"export_cap":100000,"truncated":false,"label_status":"direct model label, unvalidated","prediction_status":"machine_predicted_unvalidated (Codex and Gemma teacher distillation)","score_status":"score_only:v0-immature-baseline","snapshot":{"source":"OpenAlex, pinned release, all 482 partitions","release":"2026-06-24","frame_built":"2026-07-12"},"permalink":"https://metacan.xera.ac/q/add399e4d669","api":"https://metacan.xera.ac/api/v1/cohort?venue=Linguistic+Data+Consortium+Catalog"},"results":[{"id":"W3186788573","doi":"10.35111/srvm-p675","title":"LORELEI Akan Representative Language Pack","year":2021,"lang":"en","type":"dataset","venue":"Linguistic Data Consortium Catalog","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"route_ca_aff":false,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":true,"ca_institutions":"","funders":"","keywords":"Computer science; Annotation; Natural language processing; XML; Artificial intelligence; Context (archaeology); Resource (disambiguation); Sentence; Lemmatisation; Information retrieval; World Wide Web","score_opus":0.029608894861936852,"score_gpt":0.33350167127389563,"score_spread":0.3038927764119588,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W3186788573","genre_codex":"other","genre_gemma":"dataset","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"dataset","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.013835522,0.001617411,0.0464054,0.00488804,0.0019954778,0.0016820387,0.38727912,0.083443016,0.45885405],"genre_scores_gemma":[0.032997478,0.0011992311,0.06673688,0.0015048214,0.00062933244,0.002793233,0.5173878,0.019367803,0.35738343],"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","domain_scores_codex":[0.99720037,0.00061595556,0.00029638378,0.0006403611,0.0010290426,0.00021795461],"domain_scores_gemma":[0.98885274,0.0023932767,0.00057698047,0.0025970028,0.00503947,0.0005403791],"candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0030450919,0.0013736639,0.00093771704,0.005037313,0.001521139,0.0043932605,0.0020052572,0.0007898976,0.30193996],"category_scores_gemma":[0.009225233,0.0009789636,0.00051587704,0.0051115826,0.00072462065,0.005877991,0.004186717,0.0016894428,0.2748518],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00023474619,0.000044986104,0.0007070818,0.00044021208,0.000009297602,0.00018775825,0.0006234452,0.00022127843,0.003292623,0.0032436457,0.89637405,0.09462098],"study_design_scores_gemma":[0.0000143151165,0.000015808198,0.0009552281,0.00007740709,0.0000038153316,0.00013577816,0.00029022578,0.0004418933,0.0020560524,0.0007523025,0.995233,0.000024197123],"about_ca_topic_score_codex":0.0132698165,"about_ca_topic_score_gemma":0.012768554,"teacher_disagreement_score":0.30193996,"about_ca_system_score_codex":0.0018648144,"about_ca_system_score_gemma":0.0027931957,"threshold_uncertainty_score":0.9956979},"labels":[],"label_agreement":null},{"id":"W7104390420","doi":"10.35111/5b6x-2e75","title":"LORELEI Ilocano Incident Language Pack","year":2025,"lang":"","type":"dataset","venue":"Linguistic Data Consortium Catalog","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"route_ca_aff":false,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":true,"ca_institutions":"","funders":"","keywords":"Context (archaeology); Resource (disambiguation); Natural language; Language technology; Language identification; Event (particle physics); Information extraction","score_opus":0.026925445697423278,"score_gpt":0.3228691225035973,"score_spread":0.295943676806174,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W7104390420","genre_codex":"dataset","genre_gemma":"dataset","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"dataset","genre_consensus":"dataset","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.04298235,0.0005192096,0.07597896,0.0069852867,0.0025049138,0.008123074,0.38517645,0.114115946,0.36361375],"genre_scores_gemma":[0.06602241,0.00034365125,0.10084351,0.002721413,0.00035253773,0.0065218997,0.6870985,0.013735784,0.12236029],"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","domain_scores_codex":[0.99636614,0.0009734045,0.00030702938,0.00047607726,0.0016034513,0.00027379076],"domain_scores_gemma":[0.98975825,0.0019348153,0.00034069025,0.0017120007,0.0057064346,0.00054785877],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0050458736,0.0010588922,0.00072614703,0.0023358248,0.0016099043,0.003001187,0.0022849904,0.0010267656,0.12146631],"category_scores_gemma":[0.0122257285,0.00082224136,0.00048626907,0.0016595395,0.000844127,0.0043714484,0.003992337,0.0023309726,0.08633213],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00048400095,0.00018781626,0.0012226013,0.0005331404,0.00001183162,0.0002057952,0.00097573927,0.0006164578,0.0047591347,0.0037195585,0.9200239,0.06725996],"study_design_scores_gemma":[0.00013148737,0.00015337448,0.00279139,0.00015680245,0.000012774328,0.00023019318,0.00139148,0.0029908898,0.009612032,0.0019019948,0.9805377,0.000089875415],"about_ca_topic_score_codex":0.019081654,"about_ca_topic_score_gemma":0.018804835,"teacher_disagreement_score":0.12146631,"about_ca_system_score_codex":0.002195725,"about_ca_system_score_gemma":0.003419498,"threshold_uncertainty_score":0.4063453},"labels":[],"label_agreement":null}]}