{"id":"W7148418289","doi":"10.1109/asru65441.2025.11434627","title":"Low-Resource Domain Adaptation for Speech LLMs via Text-Only Fine-Tuning","year":2025,"lang":"","type":"article","venue":"","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Optech (Canada)","funders":"","keywords":"Adaptation (eye); Domain (mathematical analysis); Generalization; Domain adaptation; Encoder; Mechanism (biology); Degradation (telecommunications)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008736574,0.001448335,0.00111588,0.0006197749,0.0003581985,0.0007411049,0.001188604,0.0009063036,0.004301295],"category_scores_gemma":[0.003686847,0.0003300155,0.0007036143,0.0005512801,0.000614654,0.001236088,0.001583784,0.00163743,0.004100609],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004043092,"about_ca_system_score_gemma":0.000721983,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00243093,"about_ca_topic_score_gemma":0.005062277,"domain_scores_codex":[0.9993031,0.0001737897,0.00004732633,0.0002626908,0.0001433705,0.00006966738],"domain_scores_gemma":[0.9990526,0.0004198256,0.00005436071,0.0002239555,0.000188627,0.00006072691],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0005746156,0.0003767557,0.001051998,0.0002207233,0.00009419008,0.0002799482,0.0001722944,0.1258726,0.1047393,0.002567136,0.01006595,0.7539845],"study_design_scores_gemma":[0.0000682647,0.0001280293,0.0008185509,0.00002178285,0.00003644956,0.0002086923,0.00008641169,0.9471012,0.04096219,0.004881581,0.005646609,0.00004034653],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03311434,0.00105683,0.9521017,0.0002164892,0.0002346586,0.000115258,0.0002932204,0.009918259,0.002949308],"genre_scores_gemma":[0.6021829,0.0006415938,0.3794839,0.001075469,0.0002385084,0.0004709554,0.002404549,0.001414115,0.01208792],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004301295,"threshold_uncertainty_score":0.01438922,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02317349662740988,"score_gpt":0.2601340969287989,"score_spread":0.236960600301389,"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."}}