{"id":"W7126425016","doi":"10.18653/v1/2024.findings-eacl.128","title":"Jigsaw Pieces of Meaning: Modeling Discourse Coherence with Informed Negative Sample Synthesis","year":2024,"lang":"","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Atomic Energy of Canada Limited","keywords":"Coherence (philosophical gambling strategy); Jigsaw; Sample (material); Set (abstract data type)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002160763,0.0009751892,0.0006752477,0.0007047823,0.0006393441,0.00147226,0.001298636,0.001014103,0.002678766],"category_scores_gemma":[0.01248145,0.0005245603,0.0007670844,0.0004779313,0.001426543,0.002891566,0.001915327,0.001866793,0.0006805051],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007807211,"about_ca_system_score_gemma":0.0007448884,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002619457,"about_ca_topic_score_gemma":0.004929214,"domain_scores_codex":[0.9990098,0.0004783777,0.00003693591,0.0002896595,0.0001299771,0.00005524826],"domain_scores_gemma":[0.9964077,0.002533171,0.0002058592,0.0004188045,0.000333787,0.0001007417],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001156183,0.0002922767,0.006293648,0.0005377142,0.000209,0.0004919418,0.004195923,0.4408006,0.04135998,0.06895843,0.006209556,0.4294947],"study_design_scores_gemma":[0.00004138792,0.00008413364,0.0005523716,0.00002686355,0.00003349594,0.00006141861,0.0001818013,0.9473521,0.006313117,0.04275391,0.002580085,0.00001930814],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09789503,0.0005787179,0.8960568,0.0004583563,0.00007864863,0.00009976976,0.0002786249,0.001533683,0.003020319],"genre_scores_gemma":[0.7779164,0.0001606112,0.2168207,0.0002178603,0.00006423079,0.0002019673,0.0008600355,0.0005111803,0.003247051],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002678766,"threshold_uncertainty_score":0.01142734,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04639313863466781,"score_gpt":0.290176069567851,"score_spread":0.2437829309331832,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}