{"meta":{"page":1,"per_page":50,"max_per_page":100,"total":2,"total_is_capped":false,"direct_labels_cover":0,"predictions_cover":2,"direct_label_status":"direct model label, unvalidated","prediction_status":"machine_predicted_unvalidated (Codex and Gemma teacher distillation)","score_status":"score_only:v0-immature-baseline (scores rank; they never assert a category)","snapshot":{"source":"OpenAlex, pinned release, all 482 partitions","release":"2026-06-24","frame_built":"2026-07-12"},"query_hash":"918e7084900e","filters":{"venue":"Natural language processing"}},"results":[{"id":"W2548230849","doi":"10.1075/nlp.2.15mey","title":"Extracting knowledge-rich contexts for terminography","year":2001,"lang":"en","type":"book-chapter","venue":"Natural language processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":215,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Ottawa","funders":"","keywords":"Construct (python library); Computer science; Paralanguage; Domain knowledge; Domain (mathematical analysis); Knowledge extraction; Field (mathematics); Context (archaeology); Knowledge management; Natural language processing; Data science; Artificial intelligence; Psychology; Communication; Geography; Mathematics","retraction":null,"screen_n_in":null,"score":{"opus":0.01563003246822913,"gpt":0.3028685290154244,"spread":0.2872384965471953,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005306258,0.0009565764,0.0008628641,0.0008096725,0.0005663479,0.0009534777,0.002333633,0.0008533777,0.00003822714],"category_scores_gemma":[0.0002527835,0.0008446428,0.0004477105,0.0003492268,0.0001530006,0.001422905,0.00053997,0.001680204,0.00003280994],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002219155,"about_ca_system_score_gemma":0.0003137696,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000832424,"about_ca_topic_score_gemma":0.00002862515,"domain_scores_codex":[0.996295,0.00003278453,0.0007794738,0.001414361,0.0005919903,0.0008863951],"domain_scores_gemma":[0.9968852,0.0003504634,0.0009378355,0.0008897218,0.0007469194,0.0001898201],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000021406,0.00002721694,0.000003450084,0.0006022839,0.00004492642,0.0001559541,0.001538779,1.678437e-7,0.0006345861,0.01322338,0.00146435,0.9822835],"study_design_scores_gemma":[0.004744011,0.0008412653,0.00002128007,0.01851698,0.0009517972,0.003311776,0.0005417956,0.06881844,0.0217189,0.1768467,0.6901055,0.01358155],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"review","genre_gemma":"methods","genre_scores_codex":[0.00002119715,0.6957977,0.1205934,0.0002826504,0.0009542715,0.00108584,0.00002610611,0.002822111,0.1784167],"genre_scores_gemma":[0.1843339,0.0001739035,0.467833,0.0008167608,0.001707481,0.0001355013,0.0001582562,0.0003025974,0.3445386],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.968702,"threshold_uncertainty_score":0.9994004,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2475426007","doi":"10.1075/nlp.9.05hab","title":"Arabic preprocessing for Statistical Machine Translation","year":2012,"lang":"en","type":"book-chapter","venue":"Natural language processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université du Québec à Montréal","funders":"","keywords":"Arabic; Preprocessor; Computer science; Natural language processing; Translation (biology); Machine translation; Artificial intelligence; Linguistics; Philosophy; Chemistry","retraction":null,"screen_n_in":null,"score":{"opus":0.01989698859571343,"gpt":0.2977855957314866,"spread":0.2778886071357731,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0006014216,0.000811449,0.0007410979,0.0004015178,0.000428388,0.0006888921,0.0014065,0.0007106921,0.00007219503],"category_scores_gemma":[0.0001889979,0.0007015062,0.0002350892,0.0001659782,0.0001450865,0.001725386,0.0002443358,0.001338117,0.000027113],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002288124,"about_ca_system_score_gemma":0.0002748142,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001418323,"about_ca_topic_score_gemma":0.00001861665,"domain_scores_codex":[0.9964447,0.00003454072,0.0007390332,0.001178863,0.0007807054,0.0008221557],"domain_scores_gemma":[0.9977037,0.0003033603,0.0006764146,0.0007063103,0.000395898,0.0002143479],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003695383,0.00002100711,0.000001549042,0.001052808,0.00002927797,0.00002741436,0.001112626,7.270449e-7,0.0006530696,0.0168817,0.0002489543,0.9799339],"study_design_scores_gemma":[0.003468198,0.0004783465,0.00001702122,0.00755911,0.001096865,0.001046723,0.0001168683,0.6232317,0.01234051,0.1686928,0.1732036,0.008748299],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.000001046963,0.4471191,0.542789,0.0001973246,0.0002630106,0.0005337868,0.00006194871,0.001099704,0.007935077],"genre_scores_gemma":[0.1292223,0.00008761324,0.8228327,0.0004668235,0.0007937274,0.00007406979,0.0004807113,0.0001885516,0.04585355],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9711856,"threshold_uncertainty_score":0.9995436,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}