{"meta":{"query_hash":"6b3902dad86c","filters":{"venue":"Natural Language Processing Journal"},"cohort_total":14,"direct_labels_cover":0,"predictions_cover":14,"exported":14,"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/6b3902dad86c","api":"https://metacan.xera.ac/api/v1/cohort?venue=Natural+Language+Processing+Journal"},"results":[{"id":"W4389371257","doi":"10.1016/j.nlp.2023.100046","title":"Context is not key: Detecting Alzheimer’s disease with both classical and transformer-based neural language models","year":2023,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Topic Modeling","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Vector Institute; Dalhousie University; Toronto Rehabilitation Institute; University of Toronto; University Health Network","funders":"Canadian Institutes of Health Research","keywords":"Computer science; Transformer; Language model; Word2vec; Artificial intelligence; Machine learning; Artificial neural network; Natural language processing; Speech recognition","score_opus":0.028935805657169744,"score_gpt":0.2818778643010476,"score_spread":0.25294205864387787,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4389371257","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.71481377,0.019381396,0.2587133,0.006035137,0.0002637991,0.00019703379,0.000007862261,0.0005030342,0.00008467459],"genre_scores_gemma":[0.9823326,0.000009595152,0.015447871,0.0018390511,0.0002313397,0.00000677948,0.0000028716147,0.000030454357,0.000099436795],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99795675,0.00007197383,0.00031225983,0.0004596564,0.000632987,0.0005663866],"domain_scores_gemma":[0.99905425,0.00009380355,0.00016384668,0.00024298255,0.00010294914,0.00034214286],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003599578,0.00026435297,0.00025285842,0.00022161492,0.0005033079,0.00076877314,0.00053272146,0.000078625024,0.000008032755],"category_scores_gemma":[0.000051388524,0.00019057521,0.00009278945,0.00045732356,0.00007711665,0.0012526397,0.00006243584,0.0008871765,0.0000040145846],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00013577437,0.000023255612,0.00020727167,0.000071740236,0.000031204367,0.0008359195,0.014398689,0.0017973259,0.0024115243,0.000103052975,0.00004754431,0.9799367],"study_design_scores_gemma":[0.0009930996,0.00005960749,0.00043472063,0.00020151986,0.000050155028,0.00026552682,0.0011669082,0.99406147,0.0022793377,0.00015000717,0.000033249762,0.0003044202],"about_ca_topic_score_codex":0.000019235698,"about_ca_topic_score_gemma":0.000025218187,"teacher_disagreement_score":0.9922641,"about_ca_system_score_codex":0.000044488865,"about_ca_system_score_gemma":0.0002497825,"threshold_uncertainty_score":0.7771432},"labels":[],"label_agreement":null},{"id":"W4390686563","doi":"10.1016/j.nlp.2023.100052","title":"Deep temporal modelling of clinical depression through social media text","year":2024,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Mental Health via Writing","field":"Psychology","cited_by":7,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"University of Alberta","funders":"Alberta Machine Intelligence Institute","keywords":"Computer science; Timeline; Social media; Artificial intelligence; Classifier (UML); Machine learning; Depression (economics); Data mining; Statistics; World Wide Web","score_opus":0.1007322898652962,"score_gpt":0.4643236625793267,"score_spread":0.3635913727140305,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4390686563","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.68942577,0.2631169,0.031171555,0.0005110704,0.0066674855,0.00018250178,0.000009960711,0.0002217119,0.008693063],"genre_scores_gemma":[0.9796912,0.00004548838,0.016008627,0.00022371924,0.0037136583,0.0000043108284,0.000014635909,0.000049251004,0.00024908836],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9971675,0.00028265908,0.001167837,0.00035942547,0.00053813145,0.0004844415],"domain_scores_gemma":[0.9987576,0.00041719235,0.0004448983,0.00011418951,0.00012968008,0.0001364338],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001431441,0.000198895,0.00042152824,0.00013049439,0.0003468165,0.00014373449,0.0002873167,0.00029994184,0.0005528503],"category_scores_gemma":[0.00016743326,0.0001526882,0.000220747,0.00030835014,0.00012240461,0.0004522777,0.00006381647,0.0020628804,0.00004995496],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0002409965,0.00010603568,0.0022671844,0.00044787128,0.000055756154,0.0005556343,0.04420907,0.000036687055,0.00022213499,0.00021393581,0.0010717995,0.9505729],"study_design_scores_gemma":[0.011088139,0.0006010671,0.016849382,0.013828009,0.00079143146,0.011457441,0.14140266,0.7789595,0.0025669432,0.010798278,0.008633553,0.003023569],"about_ca_topic_score_codex":0.000035199835,"about_ca_topic_score_gemma":0.000012491651,"teacher_disagreement_score":0.94754934,"about_ca_system_score_codex":0.000071590024,"about_ca_system_score_gemma":0.00018126455,"threshold_uncertainty_score":0.8962304},"labels":[],"label_agreement":null},{"id":"W4396853851","doi":"10.1016/j.nlp.2024.100079","title":"Decoding depression: Analyzing social network insights for depression severity assessment with transformers and explainable AI","year":2024,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Mental Health via Writing","field":"Psychology","cited_by":13,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Queen's University","funders":"","keywords":"Depression (economics); Decoding methods; Transformer; Psychology; Computer science; Artificial intelligence; Algorithm; Engineering; Electrical engineering; Economics","score_opus":0.014749563455451602,"score_gpt":0.3798009601573698,"score_spread":0.3650513967019182,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4396853851","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.7730748,0.17242774,0.048893508,0.00085579394,0.001428072,0.000596554,0.000007831443,0.00022531346,0.0024903894],"genre_scores_gemma":[0.9847895,0.000041718366,0.012951421,0.0004122714,0.0014575594,0.000048671347,0.000015146162,0.00005025742,0.0002334495],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99805725,0.00013263038,0.00040232923,0.0004208604,0.0003218006,0.00066514144],"domain_scores_gemma":[0.999325,0.0001591332,0.00016776256,0.00006739732,0.0000981903,0.00018254797],"candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.0007179426,0.00024377815,0.00029264923,0.00015855882,0.001673733,0.0005254391,0.00013378456,0.00016887541,0.000074692085],"category_scores_gemma":[0.000016746599,0.00017326452,0.00008124099,0.00032578746,0.000057189267,0.0007398577,0.000033617867,0.0012231006,0.000001420662],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00039291865,0.00003666226,0.005728435,0.0008574909,0.000088783716,0.00037975702,0.01187622,0.000020798452,0.000763409,0.00012491844,0.0010681968,0.97866243],"study_design_scores_gemma":[0.03258547,0.0024137872,0.11912704,0.047231708,0.0027981298,0.031215407,0.15826772,0.53933173,0.006664752,0.008496994,0.04393066,0.007936617],"about_ca_topic_score_codex":0.000020066198,"about_ca_topic_score_gemma":0.00003590598,"teacher_disagreement_score":0.9707258,"about_ca_system_score_codex":0.0002178914,"about_ca_system_score_gemma":0.00017950169,"threshold_uncertainty_score":0.999626},"labels":[],"label_agreement":null},{"id":"W4401813233","doi":"10.1016/j.nlp.2024.100098","title":"HarmonyNet: Navigating hate speech detection","year":2024,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Vector Institute","funders":"Vector Institute; Government of Ontario; Canadian Institute for Advanced Research","keywords":"Voice activity detection; Computer science; Speech recognition; Speech processing","score_opus":0.006087059208944044,"score_gpt":0.26614302229227743,"score_spread":0.2600559630833334,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4401813233","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.39126688,0.093038455,0.50561935,0.0014972542,0.0056905975,0.00015408793,0.0000012141584,0.0016398622,0.0010923059],"genre_scores_gemma":[0.9700234,0.000036944883,0.028307658,0.00017610336,0.0008705454,0.0000027406882,0.0000011871149,0.000020419368,0.0005609828],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99855006,0.000059269416,0.0002597718,0.0003174104,0.00044153724,0.0003719721],"domain_scores_gemma":[0.9994756,0.00003306572,0.000102809216,0.00014667312,0.000121436446,0.00012042861],"candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.00055706885,0.00016853143,0.000123435,0.00015141189,0.00050259806,0.002464079,0.00043440162,0.0000914097,0.000019882134],"category_scores_gemma":[0.000079781996,0.00013234418,0.00010422468,0.0007051903,0.000024991039,0.0015988478,0.000073059964,0.001470094,0.00008713623],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.000004223396,0.0000072905996,0.0000030461576,0.000045616813,0.0000111544305,0.0007487012,0.001527163,0.000008311622,0.067115076,0.00002422679,0.00006689359,0.9304383],"study_design_scores_gemma":[0.00046747705,0.00013999232,0.0001311472,0.0014668463,0.000033609853,0.024363423,0.00055989804,0.60374695,0.36236247,0.0015675554,0.004568436,0.0005922083],"about_ca_topic_score_codex":0.000020333779,"about_ca_topic_score_gemma":0.000008989083,"teacher_disagreement_score":0.9298461,"about_ca_system_score_codex":0.00012727837,"about_ca_system_score_gemma":0.00009480811,"threshold_uncertainty_score":0.99857146},"labels":[],"label_agreement":null},{"id":"W4402206897","doi":"10.1016/j.nlp.2024.100102","title":"Job description parsing with explainable transformer based ensemble models to extract the technical and non-technical skills","year":2024,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Carleton University","funders":"","keywords":"Interpretability; Computer science; Margin (machine learning); Artificial intelligence; Machine learning; Transformer; Statistical model; Engineering","score_opus":0.013537905842925406,"score_gpt":0.2620085156993179,"score_spread":0.24847060985639247,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4402206897","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.060028207,0.008850482,0.92764574,0.0023953076,0.00017272773,0.00017936299,5.175306e-7,0.00018974736,0.0005378876],"genre_scores_gemma":[0.80310625,0.000012024932,0.19612667,0.0004476308,0.0001466595,0.0000105184245,4.9486044e-7,0.000016795666,0.00013293925],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.998373,0.00004820632,0.0002805291,0.00038212657,0.00049496297,0.00042121622],"domain_scores_gemma":[0.99938047,0.00008840977,0.000054423082,0.00021924489,0.000104016675,0.00015344167],"candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.00077892176,0.00019258114,0.00017463561,0.00017589511,0.0004528939,0.0016117882,0.0005038463,0.0001046594,0.0000041439016],"category_scores_gemma":[0.000032194544,0.000110847555,0.000063590414,0.0004637141,0.000049589038,0.001686995,0.00004802849,0.0010294091,0.0000023157468],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0000602988,0.000056149525,0.000035626137,0.0001650503,0.000016266427,0.00063238735,0.00671114,0.0058422857,0.083600834,0.00067999045,0.00031764468,0.90188235],"study_design_scores_gemma":[0.0003345643,0.00011758417,0.00015225718,0.0011693321,0.00003244985,0.0046327068,0.00032515472,0.987662,0.0036707653,0.001409413,0.0001949949,0.00029877454],"about_ca_topic_score_codex":0.00001290605,"about_ca_topic_score_gemma":0.000034777542,"teacher_disagreement_score":0.9818197,"about_ca_system_score_codex":0.00013828161,"about_ca_system_score_gemma":0.0002452073,"threshold_uncertainty_score":0.99942464},"labels":[],"label_agreement":null},{"id":"W4402468717","doi":"10.1016/j.nlp.2024.100105","title":"Personality and emotion—A comprehensive analysis using contextual text embeddings","year":2024,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Sentiment Analysis and Opinion Mining","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"New York Institute of Technology","funders":"","keywords":"Personality; Psychology; Natural language processing; Cognitive psychology; Computer science; Social psychology","score_opus":0.021775286337649503,"score_gpt":0.32170208357638164,"score_spread":0.2999267972387321,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4402468717","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.6990619,0.09174578,0.20797773,0.0005080463,0.00041189438,0.00004108763,0.0000020902978,0.00012187834,0.00012960278],"genre_scores_gemma":[0.97096175,0.000037478272,0.028211614,0.00023617191,0.00030077156,4.073325e-7,0.0000047387184,0.000008572143,0.00023848187],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99852574,0.000083528656,0.00030799193,0.00037312257,0.000440779,0.0002688175],"domain_scores_gemma":[0.9993187,0.000082216015,0.0001487621,0.00012213323,0.0002044186,0.00012373678],"candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003904862,0.00017295105,0.00028493511,0.0005173235,0.00045278086,0.0022731945,0.0003042941,0.00006066989,0.00007873206],"category_scores_gemma":[0.000036831123,0.00013208241,0.00022902702,0.001355565,0.00006267552,0.0011495253,0.00012050844,0.00050125824,0.0000056505046],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.000020429678,0.000053238768,0.0038742998,0.0001968081,0.0018780974,0.00047231786,0.047690902,0.0006839085,0.013550999,0.0014671452,0.0005299783,0.9295819],"study_design_scores_gemma":[0.00018314192,0.000016394182,0.0032030325,0.00017720956,0.00034223255,0.0005617214,0.0031519672,0.9915137,0.00021091345,0.00008606855,0.00033477935,0.0002188381],"about_ca_topic_score_codex":0.000022789523,"about_ca_topic_score_gemma":0.0000035444484,"teacher_disagreement_score":0.99082977,"about_ca_system_score_codex":0.00007590151,"about_ca_system_score_gemma":0.00008669513,"threshold_uncertainty_score":0.99876255},"labels":[],"label_agreement":null},{"id":"W4403096752","doi":"10.1016/j.nlp.2024.100110","title":"Recent advancements in automatic disordered speech recognition: A survey paper","year":2024,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Speech recognition; Computer science","score_opus":0.024658999476806286,"score_gpt":0.301425732541467,"score_spread":0.2767667330646607,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4403096752","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.6130421,0.32420123,0.038124293,0.0089400895,0.006828734,0.0007230544,0.000039660317,0.0014854749,0.006615342],"genre_scores_gemma":[0.88240343,0.0012522984,0.11462845,0.00093253225,0.0002439077,0.000014858368,0.00002854163,0.000028765953,0.00046721712],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9982824,0.00021009463,0.00039702817,0.00031454203,0.00045575586,0.0003401889],"domain_scores_gemma":[0.9993349,0.00015682649,0.00009874828,0.0001412547,0.00016785493,0.00010046043],"candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0009919,0.00017019249,0.00018730569,0.00032682173,0.00014898399,0.0010535192,0.0004268933,0.00007170017,0.0006661219],"category_scores_gemma":[0.00045959195,0.00013264484,0.000065405904,0.0010135297,0.000023183207,0.0016933995,0.000065626075,0.0006392855,0.00013975913],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0000073916067,0.000044451866,0.00013323396,0.000049962477,0.000011468622,0.0004339245,0.0008501826,3.3965014e-7,0.00032459298,0.0000030229482,0.00035052415,0.99779093],"study_design_scores_gemma":[0.006093896,0.0004070984,0.06204421,0.012664181,0.00011959982,0.015556719,0.004036156,0.8196458,0.0146155935,0.017663067,0.043247703,0.0039059422],"about_ca_topic_score_codex":0.000013579957,"about_ca_topic_score_gemma":0.00013192242,"teacher_disagreement_score":0.993885,"about_ca_system_score_codex":0.00014494397,"about_ca_system_score_gemma":0.00019359516,"threshold_uncertainty_score":0.9999835},"labels":[],"label_agreement":null},{"id":"W4406572044","doi":"10.1016/j.nlp.2025.100126","title":"RESPECT: A framework for promoting inclusive and respectful conversations in online communications","year":2025,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Sheridan College; Toronto Metropolitan University; Vector Institute","funders":"","keywords":"Inclusion (mineral); Psychology; Internet privacy; Sociology; Computer science; Social psychology","score_opus":0.010413841497334457,"score_gpt":0.3321838089788432,"score_spread":0.32176996748150877,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4406572044","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.19176786,0.046321336,0.72123563,0.038445134,0.000676547,0.0006336659,0.000008852952,0.00025693144,0.00065400806],"genre_scores_gemma":[0.7313312,0.000049334758,0.26806393,0.00034997336,0.00007218134,0.000007732388,0.0000027578953,0.0000046514333,0.00011816796],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99914116,0.00007583802,0.00025146818,0.00018918891,0.00013483477,0.00020748441],"domain_scores_gemma":[0.9991217,0.00024150181,0.00015030672,0.000251863,0.00018570192,0.000048913156],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004706533,0.00009419303,0.0001254723,0.00030122534,0.00060178665,0.00041643556,0.0005747952,0.000080402526,0.0000016925719],"category_scores_gemma":[0.0009778319,0.000083649546,0.00003531645,0.00064704916,0.000053186013,0.00043819303,0.00024358406,0.00075344386,5.731034e-7],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.000042390264,0.00009781494,0.0005588021,0.00008742901,0.000021418979,0.000030482317,0.019312674,0.000007891075,0.003528779,0.015903965,0.00007521842,0.9603331],"study_design_scores_gemma":[0.0040188795,0.00030576118,0.009698825,0.0047962028,0.000074404525,0.0009955489,0.014496501,0.7190985,0.00875588,0.23126675,0.0056091654,0.0008835929],"about_ca_topic_score_codex":0.00002128418,"about_ca_topic_score_gemma":0.00014618764,"teacher_disagreement_score":0.9594495,"about_ca_system_score_codex":0.00012896594,"about_ca_system_score_gemma":0.00022286983,"threshold_uncertainty_score":0.4628517},"labels":[],"label_agreement":null},{"id":"W4409360343","doi":"10.1016/j.nlp.2025.100146","title":"Detecting cognitive engagement in online course forums: A review of frameworks and methodologies","year":2025,"lang":"en","type":"review","venue":"Natural Language Processing Journal","topic":"Online and Blended Learning","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Athabasca University","funders":"Natural Sciences and Engineering Research Council of Canada; Alberta Innovates","keywords":"Course (navigation); Massive open online course; Cognition; Computer science; Psychology; Data science; World Wide Web; Engineering; Neuroscience","score_opus":0.06930139704892432,"score_gpt":0.4957217460275704,"score_spread":0.4264203489786461,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4409360343","genre_codex":"review","genre_gemma":"review","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"review","genre_consensus":"review","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.0000373909,0.9983431,0.00022919485,0.00038814786,0.00022810057,0.000406764,0.000012340845,0.000031537504,0.0003234064],"genre_scores_gemma":[0.00022632562,0.97093487,0.028020756,0.00021113387,0.00038194962,0.000009952142,0.000018170744,0.000016242375,0.000180579],"study_design_codex":"design_other","study_design_gemma":"systematic_review","domain_scores_codex":[0.9960269,0.0021154254,0.0007801683,0.0002678024,0.0004136976,0.00039600724],"domain_scores_gemma":[0.9962765,0.002203923,0.0010707964,0.00007668594,0.00030111638,0.00007098324],"candidate_categories":["metaresearch","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.0054594693,0.00026870056,0.0013483425,0.00030998242,0.00041222194,0.00012149688,0.00031324927,0.00048151423,0.000039119856],"category_scores_gemma":[0.015535945,0.00019613432,0.00024026973,0.00084509735,0.00017391024,0.00023196293,0.00013202218,0.00503028,4.1138594e-7],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0000038131757,0.00002939792,0.0000074709874,0.068717115,0.0000413239,0.000049355036,0.002092432,9.091629e-8,1.3999717e-7,0.0000069936787,0.000017558668,0.9290343],"study_design_scores_gemma":[0.00020696061,0.000037640213,0.000007107712,0.80867106,0.0010741461,0.000073025716,0.032158993,0.000012190716,7.782145e-7,0.000102473285,0.15734458,0.00031103706],"about_ca_topic_score_codex":0.000027308191,"about_ca_topic_score_gemma":0.00015803853,"teacher_disagreement_score":0.9287233,"about_ca_system_score_codex":0.0001234342,"about_ca_system_score_gemma":0.0021508848,"threshold_uncertainty_score":0.99726516},"labels":[],"label_agreement":null},{"id":"W4410109448","doi":"10.1016/j.nlp.2025.100152","title":"Bayesian Q-learning in multi-objective reward model for homophobic and transphobic text classification in low-resource languages: A hypothesis testing framework in multi-objective setting","year":2025,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"University of Calgary","funders":"Science Foundation Ireland","keywords":"Bayesian probability; Computer science; Resource (disambiguation); Machine learning; Artificial intelligence; Natural language processing; Psychology","score_opus":0.02416340836884429,"score_gpt":0.3046702726710389,"score_spread":0.2805068643021946,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4410109448","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.17828058,0.008541961,0.8112101,0.001038583,0.00005660518,0.00050887803,0.000002291735,0.00027669128,0.00008432966],"genre_scores_gemma":[0.70045334,0.000023247492,0.29920223,0.00012656036,0.000014729403,0.00006362823,0.0000014505995,0.000015872638,0.00009893471],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9978722,0.00017402714,0.0006044363,0.0006094393,0.00022312786,0.00051681494],"domain_scores_gemma":[0.99830776,0.00092772226,0.00035983013,0.00021072134,0.0001386433,0.000055310004],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009662623,0.00027490343,0.00034675866,0.0011271508,0.00031958378,0.00046952374,0.000615445,0.00028518715,9.842504e-7],"category_scores_gemma":[0.0033818546,0.00025392097,0.0000659918,0.0017586075,0.00008805827,0.0008337285,0.00009846876,0.0017106861,6.535298e-7],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00008540437,0.0002057403,0.023302726,0.00028220945,0.000015562746,0.000054970176,0.04143179,0.005699682,0.0123637235,0.00026208122,0.0000057266748,0.9162904],"study_design_scores_gemma":[0.0014132565,0.000029460494,0.031789992,0.0017161003,0.000008936712,0.00002625731,0.0155745875,0.94588107,0.001526324,0.0017511917,0.0000034089103,0.0002794345],"about_ca_topic_score_codex":0.000048362337,"about_ca_topic_score_gemma":0.0002902494,"teacher_disagreement_score":0.9401814,"about_ca_system_score_codex":0.00043569415,"about_ca_system_score_gemma":0.00027402575,"threshold_uncertainty_score":0.9999913},"labels":[],"label_agreement":null},{"id":"W4411176006","doi":"10.1016/j.nlp.2025.100159","title":"Next-generation image captioning: A survey of methodologies and emerging challenges from transformers to Multimodal Large Language Models","year":2025,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Closed captioning; Computer science; Transformer; Natural language processing; Artificial intelligence; Image (mathematics); Engineering; Electrical engineering","score_opus":0.0630583087593844,"score_gpt":0.36922933374050865,"score_spread":0.30617102498112425,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4411176006","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.36207098,0.023232965,0.61281675,0.0014575182,0.00011368827,0.00011005009,0.000011650969,0.00007837762,0.00010801007],"genre_scores_gemma":[0.69978535,0.00009196009,0.29990226,0.00011684517,0.00005051312,0.000007910132,0.000013866404,0.000008405426,0.00002289935],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9983776,0.00034414936,0.0003580232,0.00036824544,0.00026543523,0.00028650652],"domain_scores_gemma":[0.99897027,0.00028164568,0.00019345009,0.00019852059,0.000267955,0.00008817682],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0013332793,0.00018027157,0.00027342833,0.00029493222,0.00033960518,0.00034907303,0.00048527183,0.00008957508,0.0000059583817],"category_scores_gemma":[0.0006287405,0.00015364404,0.000054122134,0.00043771975,0.000038773505,0.0009760634,0.00013246156,0.00055302185,0.0000011943442],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.000026655867,0.000036442227,0.00033529187,0.000055835502,0.000038446044,0.000007667559,0.028725203,0.0012872299,0.1492496,0.0004874309,0.000020122337,0.81973004],"study_design_scores_gemma":[0.0005892712,0.000030000403,0.030684708,0.00015000871,0.000023227572,0.00002108556,0.0030431107,0.95764405,0.007010986,0.0005701382,0.000011199518,0.00022222802],"about_ca_topic_score_codex":0.0015696003,"about_ca_topic_score_gemma":0.00048466926,"teacher_disagreement_score":0.9563568,"about_ca_system_score_codex":0.00005261588,"about_ca_system_score_gemma":0.000113623704,"threshold_uncertainty_score":0.6265422},"labels":[],"label_agreement":null},{"id":"W4411442784","doi":"10.1016/j.nlp.2025.100163","title":"Can “consciousness” be observed from large language model (LLM) internal states? Dissecting LLM representations obtained from Theory of Mind test with Integrated Information Theory and Span Representation analysis","year":2025,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"World Anti-Doping Agency","funders":"","keywords":"Representation (politics); Consciousness; Test (biology); Span (engineering); Cognitive science; Psychology; Integrated information theory; Cognitive psychology; Computer science; Engineering; Structural engineering; Political science; Geology","score_opus":0.010703627425701757,"score_gpt":0.30592501771710584,"score_spread":0.2952213902914041,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4411442784","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.9623452,0.0036077115,0.032484602,0.00047893147,0.00010999628,0.00017886703,0.00023801692,0.00005542424,0.00050123956],"genre_scores_gemma":[0.99490917,0.000040734754,0.003035218,0.00022633543,0.00012461763,0.000006277685,0.00073045405,0.000009269813,0.00091794156],"study_design_codex":"qualitative","study_design_gemma":"qualitative","domain_scores_codex":[0.99794894,0.00050272903,0.00051120046,0.00025618394,0.00047671,0.0003042278],"domain_scores_gemma":[0.9978194,0.000612911,0.0006091597,0.00016693617,0.0006823743,0.000109172615],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0010193202,0.0001945372,0.00032179375,0.00031316874,0.00074720447,0.00050939684,0.00026937987,0.00012922843,0.0001520336],"category_scores_gemma":[0.002537581,0.00013717104,0.00011209605,0.0011546626,0.00019092349,0.0014451729,0.00005974541,0.0005366907,6.6253364e-7],"study_design_candidate":"qualitative","study_design_consensus":"qualitative","about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0009940087,0.00021155724,0.09087586,0.000057680838,0.0010638733,0.000057272642,0.7491104,0.0014064593,0.012898395,0.000887457,0.00014247265,0.14229454],"study_design_scores_gemma":[0.0016301308,0.00005152816,0.037735872,0.0004829759,0.0012014294,0.000008006327,0.88789004,0.060767177,0.0046063215,0.0052351607,0.000027174698,0.00036417364],"about_ca_topic_score_codex":0.010264541,"about_ca_topic_score_gemma":0.019112635,"teacher_disagreement_score":0.14193037,"about_ca_system_score_codex":0.00017165884,"about_ca_system_score_gemma":0.00028012812,"threshold_uncertainty_score":0.998786},"labels":[],"label_agreement":null},{"id":"W4413333183","doi":"10.1016/j.nlp.2025.100176","title":"Analyzing social media discourse of avian influenza outbreaks","year":2025,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"University of Guelph","funders":"Ontario Ministry of Agriculture, Food and Rural Affairs; University of Guelph","keywords":"Outbreak; Influenza A virus subtype H5N1; Social media; Virology; Sociology; Geography; Biology; Computer science; Virus; World Wide Web","score_opus":0.021597179736304994,"score_gpt":0.4121351556805344,"score_spread":0.3905379759442294,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4413333183","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.9250697,0.04182398,0.010103942,0.004633829,0.0012404033,0.0001052892,0.0000072043945,0.00009817017,0.016917499],"genre_scores_gemma":[0.993244,0.000013816475,0.005246324,0.00017350102,0.00071096915,8.245115e-7,0.0000030364893,0.000005391139,0.00060214225],"study_design_codex":"design_other","study_design_gemma":"qualitative","domain_scores_codex":[0.99855256,0.00026232874,0.00035635603,0.0001304079,0.00045612408,0.00024220691],"domain_scores_gemma":[0.9989613,0.00025474682,0.00029945298,0.000046740002,0.00036585532,0.000071895105],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0012825187,0.000099644734,0.00023965638,0.00030656342,0.0009036763,0.00019780855,0.0002942779,0.00008396762,0.00009674658],"category_scores_gemma":[0.0009901428,0.000079969126,0.00017426265,0.0008933496,0.00023940553,0.0003181213,0.00004130447,0.00045441234,0.0000017954027],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.000026325408,0.000035723955,0.0052070227,0.000026518741,0.000065941415,0.000013996661,0.04986757,0.00002639689,0.0005573566,0.0025450767,0.00015816034,0.9414699],"study_design_scores_gemma":[0.0065641417,0.000081606726,0.38444445,0.002836288,0.002545849,0.000096013784,0.44646645,0.014876608,0.003075818,0.11207701,0.024295496,0.0026402709],"about_ca_topic_score_codex":0.00012146674,"about_ca_topic_score_gemma":0.00038280062,"teacher_disagreement_score":0.93882966,"about_ca_system_score_codex":0.00009106267,"about_ca_system_score_gemma":0.00047078828,"threshold_uncertainty_score":0.6950439},"labels":[],"label_agreement":null},{"id":"W4415305391","doi":"10.1016/j.nlp.2025.100185","title":"OPT2CODE: A retrieval-augmented framework for solving linear programming problems","year":2025,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Optimization and Mathematical Programming","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Queen's University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Executable; Solver; Code generation; Domain (mathematical analysis); Benchmark (surveying); Code (set theory); Linear programming; Integer programming","score_opus":0.00951167119287442,"score_gpt":0.2937268453066785,"score_spread":0.28421517411380404,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4415305391","genre_codex":"methods","genre_gemma":"methods","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":"methods","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.0027776447,0.024647867,0.9699555,0.00036830347,0.00070353784,0.00048181962,0.0000019678812,0.0006302041,0.00043315662],"genre_scores_gemma":[0.48963574,0.00004080347,0.50942886,0.00013153927,0.00024473274,0.000021116424,0.000008776889,0.00004321531,0.00044522033],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9986677,0.000014225556,0.00044085484,0.00017671197,0.00022935227,0.0004711137],"domain_scores_gemma":[0.99932355,0.00012598177,0.00010798648,0.00011668216,0.0002193796,0.000106420506],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00042838845,0.00021348015,0.0002533871,0.00019308031,0.00035639372,0.00046000196,0.00022642786,0.00016834623,0.000025739873],"category_scores_gemma":[0.00079027977,0.00017797347,0.00012268066,0.0005479733,0.00003764365,0.0003153874,0.000035057503,0.0008261251,0.0000031737547],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00007517191,0.00009885621,0.000027544676,0.0028882811,0.00015764283,0.00002066295,0.003378899,0.009058055,0.0045876554,0.0019302386,0.00048288802,0.9772941],"study_design_scores_gemma":[0.0010416195,0.00005410292,0.000007850902,0.0028691536,0.000108515356,0.00008691769,0.0012959866,0.97955674,0.004723613,0.003818013,0.0060066218,0.00043089295],"about_ca_topic_score_codex":4.368078e-7,"about_ca_topic_score_gemma":7.9913406e-7,"teacher_disagreement_score":0.9768632,"about_ca_system_score_codex":0.000117926436,"about_ca_system_score_gemma":0.0000658723,"threshold_uncertainty_score":0.7257548},"labels":[],"label_agreement":null}]}