{"meta":{"query_hash":"34948cb5f57d","filters":{"venue":"Empirical Methods in Natural Language Processing"},"cohort_total":11,"direct_labels_cover":0,"predictions_cover":11,"exported":11,"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/34948cb5f57d","api":"https://metacan.xera.ac/api/v1/cohort?venue=Empirical+Methods+in+Natural+Language+Processing"},"results":[{"id":"W1569397218","doi":"","title":"Utilizing Extra-Sentential Context for Parsing","year":2010,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":4,"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 Toronto","funders":"","keywords":"Treebank; Parsing; Computer science; Natural language processing; Artificial intelligence; Context (archaeology); Consistency (knowledge bases); Conditional random field; Generative grammar; Set (abstract data type); Feature (linguistics); Linguistics; Programming language","score_opus":0.05448447238682705,"score_gpt":0.44656764896122564,"score_spread":0.3920831765743986,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W1569397218","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":[0.010917213,0.03161757,0.95301104,0.001395889,0.0013351659,0.0005003957,0.0000026272558,0.0009325986,0.00028752882],"genre_scores_gemma":[0.44705656,0.0000022803,0.5517959,0.00080948084,0.00019328525,0.00004418857,0.0000038277017,0.000026043428,0.000068448746],"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","domain_scores_codex":[0.9968623,0.00029411272,0.00065648975,0.00095568,0.0003851508,0.0008462884],"domain_scores_gemma":[0.99781597,0.0009231498,0.00028249342,0.00057860673,0.00023385705,0.00016593389],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0021678603,0.00039289228,0.0005089827,0.0004072031,0.00031371295,0.00058753055,0.0015336599,0.00033636985,0.000014883002],"category_scores_gemma":[0.002845242,0.000329358,0.00019339498,0.0010837193,0.0001619848,0.0012715188,0.00039302284,0.0015988022,0.0000028439954],"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":[0.000027271715,0.000049562892,0.00021877987,0.0001057461,0.000005567016,0.000035452027,0.002102491,4.562685e-7,0.13667458,0.0021109288,0.00008523027,0.8585839],"study_design_scores_gemma":[0.0022768069,0.0001603762,0.00055689336,0.000752644,0.000051452185,0.00044155814,0.0016907835,0.42622265,0.49724507,0.061255757,0.0073073204,0.0020386905],"about_ca_topic_score_codex":0.000039614362,"about_ca_topic_score_gemma":0.0000646707,"teacher_disagreement_score":0.8565452,"about_ca_system_score_codex":0.000094136376,"about_ca_system_score_gemma":0.0001674274,"threshold_uncertainty_score":0.99991584},"labels":[],"label_agreement":null},{"id":"W2117424365","doi":"","title":"Learning Noun Phrase Query Segmentation","year":2007,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":106,"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 Alberta","funders":"","keywords":"Computer science; Query expansion; Web query classification; Query language; Natural language processing; RDF query language; Information retrieval; Artificial intelligence; Precision and recall; Web search query; Noun phrase; Sargable; Query optimization; Segmentation; Phrase; Set (abstract data type); Noun; Search engine","score_opus":0.028399328598676603,"score_gpt":0.44222973249310743,"score_spread":0.4138304038944308,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W2117424365","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":[0.014914317,0.039627492,0.9425131,0.0004802801,0.00032610304,0.00022553417,3.4541395e-7,0.0011451337,0.0007676853],"genre_scores_gemma":[0.3884385,0.000006830383,0.61047983,0.0007496814,0.000113928436,0.0000102102285,0.0000049387495,0.00002083149,0.00017528259],"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","domain_scores_codex":[0.9967754,0.0005400935,0.0006450539,0.000763454,0.0005206119,0.00075538055],"domain_scores_gemma":[0.9983721,0.00066878274,0.00029788006,0.00036127493,0.00014965223,0.00015032545],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0041159405,0.00033514746,0.00037192745,0.00057490723,0.00024680592,0.00036377885,0.000992617,0.00024001837,0.000014417455],"category_scores_gemma":[0.0018967495,0.00028308076,0.000100117475,0.0020145108,0.0000894193,0.0014413442,0.00038853637,0.0014690474,0.000008771154],"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":[0.000024317567,0.000046398112,0.0013923654,0.00008092116,0.0000046044397,0.00020941152,0.0058034514,0.00000672341,0.056698587,0.00026521587,0.000028192952,0.9354398],"study_design_scores_gemma":[0.0022931208,0.00037477718,0.0050347587,0.0011834287,0.000048863294,0.00060299825,0.005862141,0.21691144,0.737013,0.025456417,0.0022430662,0.0029760331],"about_ca_topic_score_codex":0.000049588292,"about_ca_topic_score_gemma":0.000023963354,"teacher_disagreement_score":0.93246377,"about_ca_system_score_codex":0.0003093199,"about_ca_system_score_gemma":0.00011331735,"threshold_uncertainty_score":0.99996215},"labels":[],"label_agreement":null},{"id":"W2151839123","doi":"","title":"Translating Unknown Words by Analogical Learning","year":2007,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":63,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Université de Montréal","funders":"","keywords":"Lexicon; Computer science; Artificial intelligence; Natural language processing; Machine translation; Quality (philosophy); Translation (biology)","score_opus":0.030972761303358447,"score_gpt":0.432158857636875,"score_spread":0.40118609633351654,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W2151839123","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":[0.0055690715,0.089447714,0.9010648,0.00089058594,0.00019618694,0.00018061291,5.1627165e-7,0.0011998083,0.0014507056],"genre_scores_gemma":[0.44061133,0.0000053788376,0.55849576,0.0005844925,0.000086010004,0.0000067390447,0.000004749328,0.000021679862,0.00018387065],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9959212,0.0006721996,0.00080477056,0.0009609296,0.00059617864,0.0010447211],"domain_scores_gemma":[0.9980228,0.0010033845,0.00027906938,0.00036663667,0.000117738105,0.00021040023],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0048565716,0.0004220609,0.0005291502,0.0004464129,0.00029753053,0.000392742,0.001384207,0.00036268518,0.000019508476],"category_scores_gemma":[0.0020428305,0.000337727,0.00014182588,0.0026695253,0.0001359076,0.001050989,0.0003457964,0.0023003565,0.0000040051723],"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":[0.000021205442,0.000053384276,0.0010538341,0.000061568906,0.0000053931035,0.0001588641,0.0045272023,0.000005502975,0.037366774,0.0004306199,0.00004285713,0.9562728],"study_design_scores_gemma":[0.003613478,0.00066813454,0.0029924852,0.0020062602,0.00008229982,0.0008181951,0.0042030294,0.5684596,0.35232598,0.042276245,0.016797844,0.005756494],"about_ca_topic_score_codex":0.000034612982,"about_ca_topic_score_gemma":0.000025431047,"teacher_disagreement_score":0.9505163,"about_ca_system_score_codex":0.00022386278,"about_ca_system_score_gemma":0.00009313132,"threshold_uncertainty_score":0.9999075},"labels":[],"label_agreement":null},{"id":"W2186073939","doi":"","title":"Sentiment Analysis of Social Media Texts","year":2014,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Sentiment Analysis and Opinion Mining","field":"Computer Science","cited_by":2,"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":"SemEval; Computer science; Sentiment analysis; Negation; Lexicon; Task (project management); Artificial intelligence; Natural language processing; Focus (optics); Social media; Scope (computer science); World Wide Web","score_opus":0.038254256486088135,"score_gpt":0.4293544643095399,"score_spread":0.39110020782345173,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W2186073939","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":[0.08146571,0.003887133,0.91243124,0.00064824184,0.0002924187,0.000064904,9.2745586e-7,0.00008731833,0.0011221139],"genre_scores_gemma":[0.56820756,9.3691136e-7,0.4314898,0.00018292564,0.00006287398,0.0000023345783,0.000004971203,0.0000045141965,0.000044069995],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99779284,0.00058597507,0.0005040034,0.00042391024,0.00041087935,0.00028238038],"domain_scores_gemma":[0.99860966,0.0007483925,0.00026177242,0.00024304663,0.000078824836,0.000058325913],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0018137584,0.00015335463,0.00053963857,0.0007432697,0.000089129826,0.00010529,0.00056739146,0.00008619641,0.000056644152],"category_scores_gemma":[0.0005983587,0.00012303409,0.0002455947,0.0035724007,0.000053326243,0.00025692763,0.00022145882,0.00023713437,0.0000027526892],"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":[0.0000059433337,0.00006576365,0.0052335463,0.00002466891,0.00014860647,0.0000042209567,0.015174761,0.000080671554,0.0041202307,0.00072773715,0.000049993618,0.97436386],"study_design_scores_gemma":[0.0003482107,0.000013971521,0.0502922,0.000036400637,0.0002608753,0.0000011690212,0.0006169374,0.9376463,0.0094345845,0.0005876841,0.0004716423,0.0002899791],"about_ca_topic_score_codex":0.000007165239,"about_ca_topic_score_gemma":0.000012268252,"teacher_disagreement_score":0.9740739,"about_ca_system_score_codex":0.00004256527,"about_ca_system_score_gemma":0.000026333668,"threshold_uncertainty_score":0.50171846},"labels":[],"label_agreement":null},{"id":"W2252227670","doi":"","title":"Resolving “This-issue†Anaphora","year":2012,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Syntax, Semantics, Linguistic Variation","field":"Arts and Humanities","cited_by":0,"is_retracted":false,"has_abstract":false,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Anaphora (linguistics); Computer science; Natural language processing; Artificial intelligence; Resolution (logic)","score_opus":0.07225211446287715,"score_gpt":0.4151775315217796,"score_spread":0.34292541705890245,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W2252227670","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":[0.51044583,0.17791562,0.11708841,0.03939846,0.024715668,0.0017163564,0.00003433704,0.0019954205,0.12668991],"genre_scores_gemma":[0.7111987,0.000003530713,0.28468293,0.0005899086,0.0025139032,0.000010347769,0.000009310981,0.000033337008,0.00095803925],"study_design_codex":"qualitative","study_design_gemma":"qualitative","domain_scores_codex":[0.9978594,0.00043707373,0.00048340164,0.00031194792,0.00030066216,0.00060749945],"domain_scores_gemma":[0.9986997,0.00062888616,0.00019523075,0.00022598253,0.00013412428,0.00011604963],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0018364106,0.00023739926,0.0003414143,0.00025620975,0.00028829323,0.00025437516,0.00024437674,0.00011951251,0.0006019328],"category_scores_gemma":[0.0033818425,0.00019511124,0.00007790263,0.00020480811,0.00013895857,0.00050102425,0.00010724643,0.0005965248,0.00003855746],"study_design_candidate":"qualitative","study_design_consensus":"qualitative","about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00005009808,0.00014921467,0.006383854,0.00027171447,0.000028528058,0.000033838132,0.5361682,0.0000035861692,0.0018467703,0.028490562,0.00030696057,0.42626664],"study_design_scores_gemma":[0.0047425325,0.00033618716,0.03632319,0.003238111,0.00062901544,0.00019126838,0.3644049,0.07727357,0.02532265,0.17374139,0.30691913,0.00687808],"about_ca_topic_score_codex":0.00045115818,"about_ca_topic_score_gemma":0.00022375681,"teacher_disagreement_score":0.41938856,"about_ca_system_score_codex":0.00013815722,"about_ca_system_score_gemma":0.000043170574,"threshold_uncertainty_score":0.7956405},"labels":[],"label_agreement":null},{"id":"W2996912688","doi":"","title":"Discreteness in Neural Natural Language Processing","year":2019,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","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":"University of Waterloo","funders":"","keywords":"Computer science; Artificial intelligence; Artificial neural network; Focus (optics); Process (computing); Point (geometry); Space (punctuation); Natural language processing; Natural language; Machine learning; Programming language; Mathematics","score_opus":0.03199366781166351,"score_gpt":0.4167666648137828,"score_spread":0.38477299700211925,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W2996912688","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":[0.69852096,0.06723304,0.2293354,0.0015423585,0.001295279,0.0005659743,0.0000012568942,0.00046083526,0.001044897],"genre_scores_gemma":[0.6051608,0.0000021214719,0.39369708,0.00072475214,0.00011427143,0.000016331982,0.0000035939465,0.000024862955,0.00025618478],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9959178,0.0006814858,0.00074641267,0.0011494941,0.0005623692,0.0009424483],"domain_scores_gemma":[0.9984743,0.0004317101,0.00023550222,0.0006461782,0.00008440309,0.0001279321],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0016004855,0.00042584917,0.0006324568,0.0004980108,0.00011065177,0.00044490097,0.0014880344,0.00022228264,0.000022195707],"category_scores_gemma":[0.00068992726,0.00034422427,0.00012329813,0.0018946854,0.000070123046,0.0017439317,0.0006413469,0.0015538385,0.000012982665],"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":[0.000026384494,0.00004388447,0.010166295,0.00024623092,0.0000034467714,0.00017323879,0.012200564,0.000360352,0.010523439,0.00005713252,0.000003852527,0.96619517],"study_design_scores_gemma":[0.00077101775,0.00002485557,0.008908586,0.0003841771,0.000005308407,0.000068766916,0.0016374869,0.98422366,0.003016985,0.0003359645,0.000059839098,0.0005633407],"about_ca_topic_score_codex":0.000119936594,"about_ca_topic_score_gemma":0.00008135447,"teacher_disagreement_score":0.9838633,"about_ca_system_score_codex":0.00024777337,"about_ca_system_score_gemma":0.00017299222,"threshold_uncertainty_score":0.999901},"labels":[],"label_agreement":null},{"id":"W3044276988","doi":"","title":"Low-Resource Neural Network Modelling for Universal Dependency Parsing","year":2015,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"University of New Brunswick","funders":"","keywords":"Computer science; Parsing; Dependency grammar; Dependency (UML); Artificial neural network; Artificial intelligence; Resource (disambiguation); Natural language processing; Computer network","score_opus":0.07949367515106834,"score_gpt":0.41001134372687936,"score_spread":0.330517668575811,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W3044276988","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":[0.011634904,0.007694722,0.9782944,0.0012453876,0.00029233238,0.0002874044,8.492706e-7,0.00021063103,0.00033942095],"genre_scores_gemma":[0.36419046,0.0000019471468,0.634731,0.00063948386,0.0003173522,0.000018938788,0.0000039089296,0.000015338825,0.0000815909],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.997931,0.00028973407,0.00035362373,0.0005713038,0.0002456333,0.0006086953],"domain_scores_gemma":[0.9986542,0.0005914646,0.00014368916,0.00031646577,0.00010052856,0.00019364854],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0010662008,0.00021455335,0.0002828628,0.0000964081,0.00022085324,0.00023611357,0.0007874848,0.00011993768,0.0000012849425],"category_scores_gemma":[0.00016485444,0.00018444081,0.00009287571,0.0010449523,0.00004953751,0.00059262995,0.00024986453,0.00051066437,0.0000015116807],"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":[0.00003241791,0.000043288106,0.00013565243,0.000033947443,0.000004607319,0.000023256183,0.0019379419,0.20583512,0.00031964024,0.0009405983,0.00041974886,0.7902738],"study_design_scores_gemma":[0.00038233024,0.000030033554,0.000019225274,0.00007589812,0.000007615041,0.000020267056,0.00023051938,0.98852944,0.0005926422,0.008215018,0.0016447797,0.00025225076],"about_ca_topic_score_codex":0.000012830325,"about_ca_topic_score_gemma":0.0000066667653,"teacher_disagreement_score":0.79002154,"about_ca_system_score_codex":0.000107419175,"about_ca_system_score_gemma":0.00009743714,"threshold_uncertainty_score":0.75212777},"labels":[],"label_agreement":null},{"id":"W3088099248","doi":"","title":"Memory Augmented Neural Networks for Natural Language Processing","year":2017,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Multimodal Machine Learning Applications","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":"Université de Montréal","funders":"","keywords":"Computer science; Artificial intelligence; Artificial neural network; Scalability; Set (abstract data type); Focus (optics); Auxiliary memory; Deep learning; Machine learning; Theoretical computer science; Programming language","score_opus":0.040286962499015484,"score_gpt":0.46035881602918827,"score_spread":0.4200718535301728,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W3088099248","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":[0.05030762,0.027200403,0.9156943,0.0040027914,0.0008103686,0.0008172406,0.0000031278241,0.0006191671,0.00054498785],"genre_scores_gemma":[0.54358995,0.0000015123286,0.4551488,0.0006434617,0.00027280077,0.00010148536,0.000010848813,0.00003107774,0.00020009685],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9966563,0.00044075228,0.0005958211,0.001053883,0.00038218958,0.00087107345],"domain_scores_gemma":[0.99721557,0.0007166532,0.00060530717,0.0011073457,0.00017996361,0.0001751758],"candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0016896045,0.00043113212,0.000532785,0.00024382824,0.0011750803,0.0012144708,0.0025264472,0.00022194274,0.000007850549],"category_scores_gemma":[0.0028219798,0.00036142985,0.00017728223,0.00056348677,0.00019012355,0.0013422254,0.0007465034,0.0014469394,0.0000034576285],"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":[0.000035662204,0.00005911157,0.00121685,0.00012840184,0.000009368355,0.00002934985,0.0035171157,0.00073933386,0.0031065175,0.000041282135,0.0000442566,0.9910728],"study_design_scores_gemma":[0.0008557303,0.000030319809,0.018324006,0.00014432459,0.000017588629,0.00003622402,0.00030098975,0.97801214,0.0014589496,0.00022157795,0.00012551312,0.00047260275],"about_ca_topic_score_codex":0.00022173747,"about_ca_topic_score_gemma":0.00004844307,"teacher_disagreement_score":0.99060017,"about_ca_system_score_codex":0.00016528214,"about_ca_system_score_gemma":0.00010897353,"threshold_uncertainty_score":0.9998838},"labels":[],"label_agreement":null},{"id":"W3089160253","doi":"","title":"Semantic Similarity Frontiers: From Concepts to Documents","year":2015,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Semantic similarity; Similarity (geometry); Semantics (computer science); Natural language processing; Information retrieval; Sentence; SemEval; Artificial intelligence; Word (group theory); Semantic computing; Component (thermodynamics); Semantic Web; Linguistics","score_opus":0.07529857966887143,"score_gpt":0.46713590103802416,"score_spread":0.3918373213691527,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W3089160253","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":[0.048560247,0.010694468,0.9368423,0.001973785,0.0010458494,0.0001822592,0.0000012387253,0.00019465573,0.0005051718],"genre_scores_gemma":[0.34868288,9.4030156e-7,0.6490756,0.001987621,0.00012819516,0.000008671129,0.0000018167677,0.000009349011,0.000104885265],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99744755,0.00055187405,0.0003959787,0.00070593006,0.00044312843,0.00045554637],"domain_scores_gemma":[0.99881494,0.00022375416,0.00009846788,0.00048392598,0.00010163442,0.0002772476],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0012633397,0.00021448526,0.00036626594,0.00017169403,0.000069340276,0.00029085585,0.0010429064,0.0001314226,0.0000074752797],"category_scores_gemma":[0.0015944048,0.00018275011,0.000055349312,0.0007823197,0.00003590989,0.00077594956,0.00056988286,0.0005199993,0.000014713746],"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":[0.000019861818,0.000043866185,0.005573899,0.000021636974,0.000009660027,0.00009224698,0.0189201,0.00021439175,0.0011693859,0.00005360104,0.0008722548,0.9730091],"study_design_scores_gemma":[0.0012896303,0.00006346596,0.002925541,0.00025989406,0.000017708915,0.000015156806,0.0021655464,0.95837885,0.0073293624,0.023337092,0.0034280394,0.0007897062],"about_ca_topic_score_codex":0.00023150566,"about_ca_topic_score_gemma":0.000021878353,"teacher_disagreement_score":0.9722194,"about_ca_system_score_codex":0.00020552636,"about_ca_system_score_gemma":0.00013717148,"threshold_uncertainty_score":0.7452333},"labels":[],"label_agreement":null},{"id":"W3213820586","doi":"","title":"How to Select One Among All? An Extensive Empirical Study Towards the Robustness of Knowledge Distillation in Natural Language Understanding.","year":2021,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":4,"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; Queen's University","funders":"","keywords":"Robustness (evolution); Computer science; Adversarial system; Distillation; Artificial intelligence; Machine learning; Benchmark (surveying); Domain knowledge; Natural language; Artificial neural network; Theoretical computer science","score_opus":0.10184368763546528,"score_gpt":0.4440969145619636,"score_spread":0.3422532269264983,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W3213820586","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":[0.44144806,0.004485593,0.55130064,0.0017269625,0.000418104,0.0004351607,0.0000010168233,0.00010628323,0.00007819655],"genre_scores_gemma":[0.7291704,0.0000015959038,0.27025816,0.00022602009,0.00018634,0.00002286973,0.000008113249,0.000030636955,0.000095882606],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99348384,0.0033528255,0.00066837855,0.001104063,0.00070578407,0.0006851104],"domain_scores_gemma":[0.9969469,0.0015931727,0.00030983417,0.0006558951,0.00033389457,0.00016033047],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0028106864,0.00041131588,0.00075883727,0.00045163295,0.00021644341,0.00044672334,0.0011410245,0.00020717838,0.0000073253404],"category_scores_gemma":[0.0071809455,0.00030794888,0.00012007747,0.003927593,0.00015572204,0.0010758019,0.0009116746,0.001693806,6.7639047e-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":[0.00023477798,0.001084221,0.063567854,0.00028609953,0.000102250735,0.0006473039,0.21789682,0.015081902,0.0047914227,0.00019463243,0.000037354588,0.6960754],"study_design_scores_gemma":[0.001233748,0.0001870639,0.16912736,0.00034705442,0.000053611097,0.00004367035,0.04061834,0.7854391,0.0019836822,0.0002942358,0.000019116791,0.000653034],"about_ca_topic_score_codex":0.00011371948,"about_ca_topic_score_gemma":0.0010302657,"teacher_disagreement_score":0.7703572,"about_ca_system_score_codex":0.00056470226,"about_ca_system_score_gemma":0.0003197724,"threshold_uncertainty_score":0.99993724},"labels":[],"label_agreement":null},{"id":"W34986029","doi":"10.1089/cmb.2021.0438","title":"Exploiting Conversation Structure in Unsupervised Topic Segmentation for Emails","year":2010,"lang":"en","type":"article","venue":"Empirical Methods in Natural Language Processing","topic":"Personal Information Management and User Behavior","field":"Decision Sciences","cited_by":22,"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 British Columbia","funders":"","keywords":"Latent Dirichlet allocation; Computer science; Topic model; Artificial intelligence; Natural language processing; Conversation; Segmentation; Exploit; Market segmentation; Thread (computing); Text segmentation; Information retrieval; Linguistics","score_opus":0.2587489947827946,"score_gpt":0.5735565170127669,"score_spread":0.31480752222997227,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W34986029","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":[0.9556915,0.00028552438,0.041982986,0.000927624,0.00048773555,0.00040096205,0.000005300035,0.00003840705,0.00017993372],"genre_scores_gemma":[0.6626474,5.192638e-7,0.33657724,0.00051347964,0.000078666606,0.000024668696,0.000017891245,0.0000060184216,0.0001341133],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9980823,0.00021915205,0.00064632273,0.00032102386,0.00048001143,0.0002512028],"domain_scores_gemma":[0.99848115,0.00095675973,0.00021048402,0.00016220524,0.00014093642,0.000048438582],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0026746204,0.00013636939,0.00023033822,0.00047195566,0.0000935359,0.00030318846,0.00033779393,0.00012017481,0.00019468983],"category_scores_gemma":[0.0033593937,0.00010343563,0.000061978506,0.00091790425,0.000038431328,0.0011884575,0.000062094674,0.00042272883,0.0000043555933],"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":[0.0000345721,0.000016600121,0.034156412,0.00003473535,0.000001615821,0.000003212827,0.015103693,0.000024109358,0.058859695,0.000067139765,0.000055877343,0.89164233],"study_design_scores_gemma":[0.0057115625,0.000095124604,0.39525715,0.00019011059,0.000045662156,0.00001544265,0.09290363,0.39581904,0.08027226,0.025160654,0.0033347595,0.0011946277],"about_ca_topic_score_codex":0.000018322,"about_ca_topic_score_gemma":0.00018853263,"teacher_disagreement_score":0.89044774,"about_ca_system_score_codex":0.000059386737,"about_ca_system_score_gemma":0.000041772106,"threshold_uncertainty_score":0.42179826},"labels":[],"label_agreement":null}]}