{"meta":{"query_hash":"20522714e91c","filters":{"venue":"Findings of the Association for Computational Linguistics: ACL 2022"},"cohort_total":9,"direct_labels_cover":0,"predictions_cover":9,"exported":9,"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/20522714e91c","api":"https://metacan.xera.ac/api/v1/cohort?venue=Findings+of+the+Association+for+Computational+Linguistics%3A+ACL+2022"},"results":[{"id":"W3205231356","doi":"10.18653/v1/2022.findings-acl.135","title":"Morphosyntactic Tagging with Pre-trained Language Models for Arabic and its Dialects","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"route_ca_aff":false,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"","funders":"York University; New York University Abu Dhabi","keywords":"Transformer; Arabic; Computer science; Modern Standard Arabic; Natural language processing; Artificial intelligence; Language model; Training set; Resource (disambiguation); Linguistics; Engineering; Voltage","score_opus":0.008924129542520672,"score_gpt":0.25229433331955636,"score_spread":0.2433702037770357,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W3205231356","genre_codex":"methods","genre_gemma":"methods","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"methods","genre_consensus":"methods","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.42444018,0.004453862,0.4374241,0.0011931247,0.0014773089,0.00043995076,0.015982319,0.091139264,0.02344993],"genre_scores_gemma":[0.6342845,0.0012259,0.3018605,0.0006494557,0.0001262914,0.0002666144,0.046810754,0.003675458,0.0111005455],"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","domain_scores_codex":[0.99890196,0.000272595,0.00008983855,0.00051294436,0.00012330967,0.00009931322],"domain_scores_gemma":[0.9966467,0.0016506803,0.00011110572,0.00076817663,0.0007044837,0.00011883266],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0015900575,0.0024461974,0.0008345883,0.0015592136,0.001009802,0.0023421014,0.0018375908,0.0012216264,0.006922849],"category_scores_gemma":[0.006235406,0.0007230107,0.0012335207,0.0012926352,0.00064310094,0.0035100887,0.0019128477,0.002592636,0.009834396],"study_design_candidate":"bench_or_experimental","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":[0.0010802882,0.00053900864,0.01312763,0.0010274561,0.000710432,0.00091757823,0.0014145594,0.10359788,0.057953253,0.003359026,0.039569054,0.77670383],"study_design_scores_gemma":[0.00015914497,0.00032363157,0.0084121395,0.0002231477,0.00045676986,0.0008174787,0.0012915544,0.8382786,0.09793703,0.010123746,0.041729093,0.0002476256],"about_ca_topic_score_codex":0.014101338,"about_ca_topic_score_gemma":0.021663705,"teacher_disagreement_score":0.014101338,"about_ca_system_score_codex":0.0012892233,"about_ca_system_score_gemma":0.0013878378,"threshold_uncertainty_score":0.028038502},"labels":[],"label_agreement":null},{"id":"W3206786886","doi":"10.18653/v1/2022.findings-acl.316","title":"Zero-Shot Dense Retrieval with Momentum Adversarial Domain Invariant Representations","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":19,"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; Invariant (physics); Classifier (UML); Source code; Embedding; Adversarial system; Artificial intelligence; Encoder; Autoencoder; Theoretical computer science; Pattern recognition (psychology); Algorithm; Deep learning; Mathematics","score_opus":0.01593641139419208,"score_gpt":0.2523305473785822,"score_spread":0.2363941359843901,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W3206786886","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.057948958,0.00092972006,0.93409723,0.00046163588,0.00008856666,0.00014724434,0.0003433411,0.003039991,0.0029432864],"genre_scores_gemma":[0.83946186,0.0003885376,0.14939755,0.0006139449,0.00013133713,0.00017615138,0.0016657007,0.00021887654,0.007946129],"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9991598,0.00027889566,0.000035783243,0.00025140942,0.00017796563,0.00009615626],"domain_scores_gemma":[0.9985297,0.00074993167,0.00010437807,0.00042788277,0.00012322505,0.00006493912],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0017181082,0.0010942911,0.0013653791,0.0006287016,0.000369049,0.00090734044,0.0022490679,0.0014400224,0.0019105269],"category_scores_gemma":[0.0047670393,0.0004048763,0.0007021881,0.00067443214,0.0010969882,0.0029658363,0.0018357806,0.0018612213,0.00096905883],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00037038221,0.00029921095,0.0011855,0.00019357886,0.00012251531,0.00017497834,0.000118317934,0.7372962,0.0084207915,0.02093991,0.009501789,0.2213767],"study_design_scores_gemma":[0.000012134371,0.00005767493,0.00010168358,0.0000051149395,0.0000071260033,0.000040902796,0.000008968065,0.9915775,0.0013951027,0.006324028,0.00046159507,0.000008216227],"about_ca_topic_score_codex":0.003816705,"about_ca_topic_score_gemma":0.004213269,"teacher_disagreement_score":0.003816705,"about_ca_system_score_codex":0.0009377407,"about_ca_system_score_gemma":0.00079372455,"threshold_uncertainty_score":0.009086311},"labels":[],"label_agreement":null},{"id":"W4226145103","doi":"10.18653/v1/2022.findings-acl.326","title":"On the data requirements of probing","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Topic Modeling","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":"Vector Institute; University of Toronto","funders":"","keywords":"Computer science; Reliability (semiconductor); Context (archaeology); Construct (python library); Data mining; Machine learning; Artificial intelligence; Power (physics)","score_opus":0.054162738601187525,"score_gpt":0.291028757242492,"score_spread":0.23686601864130447,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4226145103","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.12935854,0.007681459,0.6953404,0.11801296,0.0022876658,0.0019412522,0.012944842,0.006339601,0.026093261],"genre_scores_gemma":[0.50664765,0.0023372394,0.44610903,0.013650176,0.0023368895,0.0041142846,0.01786789,0.0027051382,0.004231726],"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.7993628,0.14663368,0.010859938,0.017439358,0.022870181,0.0028341473],"domain_scores_gemma":[0.14934397,0.7665843,0.006776336,0.059332296,0.014480481,0.0034826177],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.15146323,0.0020775236,0.005660926,0.003211724,0.004348049,0.011224856,0.01005458,0.0107450625,0.011888778],"category_scores_gemma":[0.685493,0.0037523624,0.0031971177,0.007854061,0.010096867,0.034245517,0.014305756,0.014496589,0.0047530113],"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":[0.009351456,0.0011847921,0.054059166,0.0027866417,0.0007923673,0.0015822558,0.006661666,0.066939525,0.008518392,0.44857207,0.12005045,0.27950126],"study_design_scores_gemma":[0.0008555682,0.00048458087,0.0069303177,0.00051169546,0.00019131607,0.0012455896,0.0024988814,0.23806246,0.003625033,0.719126,0.02631887,0.00014966173],"about_ca_topic_score_codex":0.0046378314,"about_ca_topic_score_gemma":0.0035095837,"teacher_disagreement_score":0.15146323,"about_ca_system_score_codex":0.0037036038,"about_ca_system_score_gemma":0.006071488,"threshold_uncertainty_score":0.80102366},"labels":[],"label_agreement":null},{"id":"W4280645526","doi":"10.18653/v1/2022.findings-acl.164","title":"Richer Countries and Richer Representations","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":8,"is_retracted":false,"has_abstract":true,"route_ca_aff":false,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; National Science Foundation","keywords":"Embedding; Space (punctuation); Vocabulary; Inequality; Computer science; Word lists by frequency; Word (group theory); Power (physics); Work (physics); Natural language processing; Artificial intelligence; Econometrics; Linguistics; Economics; Mathematics; Engineering","score_opus":0.011301511035746859,"score_gpt":0.2913571103954506,"score_spread":0.28005559935970376,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4280645526","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.9346654,0.00049277383,0.03767307,0.0022332792,0.0000542475,0.000025004194,0.0013786077,0.00024674702,0.023230916],"genre_scores_gemma":[0.9942263,0.000108121865,0.0040694904,0.00011364124,0.000014492049,0.000011509177,0.0006308952,0.00004233015,0.00078325125],"study_design_codex":"observational","study_design_gemma":"observational","domain_scores_codex":[0.99738544,0.0012680972,0.00019001632,0.00060028053,0.0002243815,0.00033182377],"domain_scores_gemma":[0.98621273,0.006931038,0.0021340488,0.0035969121,0.000657066,0.00046814856],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0024202068,0.0005081706,0.00047439593,0.0018805241,0.0012125312,0.00400865,0.000467019,0.00068904366,0.013380128],"category_scores_gemma":[0.02420524,0.00033758138,0.0005194135,0.0024041862,0.0020024136,0.006975264,0.004294997,0.001118591,0.00090072857],"study_design_candidate":"observational","study_design_consensus":"observational","about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0010703041,0.00026143284,0.4152344,0.00063389895,0.0008294757,0.00094059686,0.029701069,0.02700936,0.014040264,0.241716,0.008277794,0.2602854],"study_design_scores_gemma":[0.000117047595,0.0005061375,0.29241884,0.00055844226,0.00059563963,0.0017859571,0.03369395,0.0667417,0.009463291,0.5344805,0.059410676,0.00022778858],"about_ca_topic_score_codex":0.002906518,"about_ca_topic_score_gemma":0.0035231693,"teacher_disagreement_score":0.013380128,"about_ca_system_score_codex":0.0007200519,"about_ca_system_score_gemma":0.00041264683,"threshold_uncertainty_score":0.044761002},"labels":[],"label_agreement":null},{"id":"W4281263937","doi":"10.18653/v1/2022.findings-acl.117","title":"Two-Step Question Retrieval for Open-Domain QA","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"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":"Ministry of Science and ICT, South Korea; National Research Foundation of Korea; National Research Foundation","keywords":"Search engine indexing; Inference; Computer science; Information retrieval; Labrador Retriever; Pipeline (software); Squid; Domain (mathematical analysis); Artificial intelligence; Open domain; Question answering; Mathematics; Programming language","score_opus":0.01875750815666395,"score_gpt":0.2925474942045004,"score_spread":0.27378998604783644,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4281263937","genre_codex":"methods","genre_gemma":"methods","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"methods","genre_consensus":"methods","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.027755762,0.0010419132,0.95185965,0.0008017536,0.0001089852,0.00037247493,0.0006811127,0.014229616,0.0031487243],"genre_scores_gemma":[0.48502052,0.00048530745,0.5036844,0.0005585844,0.00016379434,0.0003594253,0.0033241517,0.00045971698,0.005944135],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9985671,0.00059229607,0.00008777731,0.00040124197,0.0002341387,0.000117505864],"domain_scores_gemma":[0.99486226,0.0026822905,0.00016620637,0.0013664119,0.00071025547,0.00021262499],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0034847767,0.000825689,0.0010310212,0.0013334843,0.00067610055,0.0011986323,0.002568392,0.0017909621,0.009398755],"category_scores_gemma":[0.010890025,0.00066473644,0.0012298697,0.0009476984,0.0009235206,0.005374018,0.002892502,0.0028009457,0.005467695],"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":[0.001002704,0.0008574103,0.0052059474,0.00095487264,0.00018287523,0.0002082218,0.00086227385,0.07719907,0.025425572,0.028891327,0.03443011,0.82477957],"study_design_scores_gemma":[0.00008736369,0.00019659585,0.00081500714,0.000021865451,0.000036723184,0.00019679489,0.00007878431,0.96172875,0.010281356,0.020159071,0.006366363,0.000031284897],"about_ca_topic_score_codex":0.008352006,"about_ca_topic_score_gemma":0.010537689,"teacher_disagreement_score":0.009398755,"about_ca_system_score_codex":0.0012339873,"about_ca_system_score_gemma":0.0021475325,"threshold_uncertainty_score":0.031441987},"labels":[],"label_agreement":null},{"id":"W4285148079","doi":"10.18653/v1/2022.findings-acl.293","title":"Local Structure Matters Most: Perturbation Study in NLU","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","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":"Canadian Institute for Advanced Research; McGill University; Polytechnique Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"Canada First Research Excellence Fund; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Perturbation (astronomy); Word order; Phenomenon; Artificial neural network; Invariant (physics); Artificial intelligence; Natural language processing; Mathematics; Physics","score_opus":0.010150799814172976,"score_gpt":0.24417974456153907,"score_spread":0.2340289447473661,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4285148079","genre_codex":"empirical","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"empirical","genre_consensus":"empirical","domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.9346654,0.00047688815,0.06195209,0.0005671018,0.000027730279,0.000046848007,0.00007889848,0.0003223946,0.0018626552],"genre_scores_gemma":[0.9949645,0.00006364839,0.004639853,0.000049752696,0.000012910659,0.000019638363,0.000051909425,0.00004494576,0.00015274045],"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","domain_scores_codex":[0.99860185,0.00091968477,0.000047899954,0.00023708747,0.00014227604,0.000051316518],"domain_scores_gemma":[0.9796521,0.016794078,0.0008062698,0.0019396228,0.00049690896,0.00031100394],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002466983,0.00034083024,0.00057369564,0.00044032556,0.00041083485,0.0009341705,0.00060907303,0.0007648727,0.0010041981],"category_scores_gemma":[0.033530142,0.0002995041,0.00030692236,0.00042781266,0.0012964723,0.0027097592,0.00093932706,0.0016189405,0.00018933485],"study_design_candidate":"bench_or_experimental","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":[0.0027328664,0.0010644355,0.0444022,0.00053365837,0.0004686507,0.0006758727,0.0040427293,0.6395189,0.14533497,0.03391996,0.0022522218,0.12505351],"study_design_scores_gemma":[0.00004116764,0.00029010462,0.014469592,0.000021912398,0.000048947168,0.000108375025,0.000371936,0.9240426,0.020874677,0.03900556,0.0006899219,0.000035261513],"about_ca_topic_score_codex":0.0015410637,"about_ca_topic_score_gemma":0.0011587379,"teacher_disagreement_score":0.002466983,"about_ca_system_score_codex":0.00064279296,"about_ca_system_score_gemma":0.00028849972,"threshold_uncertainty_score":0.013046801},"labels":[],"label_agreement":null},{"id":"W4285152239","doi":"10.18653/v1/2022.findings-acl.225","title":"Extracting Person Names from User Generated Text: Named-Entity Recognition for Combating Human Trafficking","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"route_ca_aff":false,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"","funders":"Samsung; Institut de Valorisation des Données; Canadian Institute for Advanced Research","keywords":"Computer science; Named-entity recognition; Punctuation; Natural language processing; Domain (mathematical analysis); Task (project management); Artificial intelligence; Grammar; Named entity; Information retrieval; Linguistics","score_opus":0.043147769997126245,"score_gpt":0.2892943644149413,"score_spread":0.24614659441781503,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4285152239","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.37594676,0.004301409,0.5652753,0.0026229264,0.0011438204,0.00051891693,0.013328119,0.025309065,0.011553624],"genre_scores_gemma":[0.608487,0.0017022827,0.3499383,0.0006134751,0.00033432714,0.00018592247,0.028475657,0.00046636513,0.009796586],"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","domain_scores_codex":[0.99898964,0.00035849298,0.00010600965,0.00028963204,0.0001985506,0.000057741363],"domain_scores_gemma":[0.9975261,0.0012457813,0.0003123847,0.00048039085,0.00036997674,0.000065389155],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0011954852,0.0007048884,0.00043148015,0.0019002715,0.0004867804,0.0007049469,0.00077037205,0.0011460476,0.0016213144],"category_scores_gemma":[0.003513259,0.00015565979,0.00046755216,0.0012269188,0.00051812764,0.0025512741,0.00097050756,0.00072638923,0.0034185234],"study_design_candidate":"bench_or_experimental","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":[0.0005261494,0.0003688666,0.01103248,0.0008043283,0.00011061259,0.0016795695,0.0007570369,0.017118024,0.046683125,0.0034318285,0.047974415,0.8695135],"study_design_scores_gemma":[0.000070349015,0.0003652453,0.021536881,0.00016998596,0.00018387806,0.0036339506,0.0016306229,0.6872189,0.17541632,0.011433456,0.09819223,0.00014813153],"about_ca_topic_score_codex":0.0011487112,"about_ca_topic_score_gemma":0.0018781663,"teacher_disagreement_score":0.0019002715,"about_ca_system_score_codex":0.00031013024,"about_ca_system_score_gemma":0.0004459386,"threshold_uncertainty_score":0.0063224435},"labels":[],"label_agreement":null},{"id":"W4285155190","doi":"10.18653/v1/2022.findings-acl.168","title":"Question Generation for Reading Comprehension Assessment by Modeling How and What to Ask","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Topic Modeling","field":"Computer Science","cited_by":21,"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":"Reading comprehension; Computer science; Comprehension; Reading (process); Focus (optics); Artificial intelligence; Natural language processing; Mathematics education; Linguistics; Psychology","score_opus":0.02239636680911099,"score_gpt":0.2829373594837503,"score_spread":0.2605409926746393,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4285155190","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.22219907,0.0064944634,0.60186166,0.004943758,0.0007422758,0.0030653747,0.08773612,0.05824472,0.014712581],"genre_scores_gemma":[0.39236644,0.0007003239,0.41570556,0.0011673897,0.000195863,0.002426306,0.18120879,0.00087994774,0.0053494032],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9968746,0.0016660356,0.00022254042,0.000853925,0.0002914026,0.00009143002],"domain_scores_gemma":[0.9868009,0.009019402,0.0005819158,0.0019199513,0.0013731784,0.0003045892],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0032771386,0.0019598466,0.00061732094,0.002861357,0.0006599366,0.0016753619,0.0027680222,0.0030816114,0.005406877],"category_scores_gemma":[0.022015857,0.00044099687,0.0018138688,0.0015028715,0.0006542341,0.004015435,0.0018446759,0.003114311,0.0051353485],"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":[0.0013665336,0.0023153657,0.06082597,0.0033944335,0.0006029955,0.0006955085,0.0031235549,0.07085049,0.022492453,0.009141075,0.13241144,0.69278026],"study_design_scores_gemma":[0.0003295262,0.00075559574,0.023405893,0.00032725005,0.00026007328,0.0006765441,0.0009786909,0.8290964,0.023978934,0.031021897,0.08902275,0.00014643806],"about_ca_topic_score_codex":0.0063370056,"about_ca_topic_score_gemma":0.014286781,"teacher_disagreement_score":0.0063370056,"about_ca_system_score_codex":0.0014490603,"about_ca_system_score_gemma":0.001316939,"threshold_uncertainty_score":0.018087864},"labels":[],"label_agreement":null},{"id":"W4285255856","doi":"10.18653/v1/2022.findings-acl.177","title":"ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Data Visualization and Analytics","field":"Computer Science","cited_by":246,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"York University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Question answering; Benchmark (surveying); Vocabulary; Chart; Visual reasoning; Artificial intelligence; Qualitative reasoning; Variety (cybernetics); Natural language processing; Linguistics","score_opus":0.00984609113906177,"score_gpt":0.2772540827683285,"score_spread":0.26740799162926676,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4285255856","genre_codex":"dataset","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"metacan-v3-hybrid-931329e0061c","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.10327149,0.01339643,0.19335051,0.0064070844,0.0014497095,0.0042034723,0.4923486,0.14562574,0.039946914],"genre_scores_gemma":[0.14033036,0.0021525414,0.26647303,0.0013529354,0.00025840412,0.0022462502,0.5780025,0.0026944452,0.00648953],"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","domain_scores_codex":[0.9921536,0.0026662683,0.0011325782,0.0016378715,0.002019465,0.0003902448],"domain_scores_gemma":[0.9761578,0.015785413,0.0010556658,0.0024764475,0.003621586,0.00090313656],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004111089,0.0032125623,0.0011379914,0.0063949833,0.0012699497,0.0032008183,0.0038516603,0.0032558877,0.013909892],"category_scores_gemma":[0.04093298,0.0005008758,0.0024689927,0.005371743,0.0010489743,0.0062629306,0.0032562187,0.00258332,0.0068232664],"study_design_candidate":"bench_or_experimental","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":[0.0011525395,0.0012425005,0.011474611,0.009581684,0.00034361662,0.0006114886,0.0016837401,0.03366618,0.008763603,0.017143106,0.65396464,0.2603723],"study_design_scores_gemma":[0.00082328205,0.0008586244,0.017402446,0.0013273364,0.00023167123,0.0010293314,0.0029545745,0.38338608,0.022382243,0.060139954,0.5092036,0.00026078118],"about_ca_topic_score_codex":0.025359798,"about_ca_topic_score_gemma":0.029044537,"teacher_disagreement_score":0.025359798,"about_ca_system_score_codex":0.0027867905,"about_ca_system_score_gemma":0.0036543335,"threshold_uncertainty_score":0.050424397},"labels":[],"label_agreement":null}]}