{"id":"W4389524537","doi":"10.18653/v1/2023.findings-emnlp.822","title":"Efficiently Enhancing Zero-Shot Performance of Instruction Following Model via Retrieval of Soft Prompt","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science and ICT, South Korea; Korea Advanced Institute of Science and Technology","keywords":"Computer science; Embedding; Benchmark (surveying); Generalization; Task (project management); Inference; Zero (linguistics); Shot (pellet); Artificial intelligence; Scaling; Machine learning; Mathematics","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":[0.002005443,0.002313481,0.001572701,0.0005472077,0.0005122755,0.001698599,0.002635477,0.002093351,0.005536102],"category_scores_gemma":[0.01047991,0.0005432037,0.001120448,0.0005152895,0.0007840086,0.004887613,0.00251721,0.003642032,0.003538924],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009378787,"about_ca_system_score_gemma":0.001587298,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005598349,"about_ca_topic_score_gemma":0.007919432,"domain_scores_codex":[0.9990358,0.0002421131,0.00005284311,0.0004156397,0.0001076037,0.0001459244],"domain_scores_gemma":[0.9973605,0.001372945,0.00009882163,0.0007041973,0.0002680011,0.0001954835],"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":[0.002039659,0.001095699,0.008909847,0.0008733749,0.0002663605,0.0003502318,0.000548374,0.1950189,0.02812785,0.00437127,0.02669523,0.7317032],"study_design_scores_gemma":[0.00007908564,0.0004635651,0.001199317,0.0000487683,0.00005749434,0.000123039,0.000227986,0.9743978,0.0124178,0.008624912,0.00231741,0.00004287294],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4137396,0.005152839,0.5119793,0.001668305,0.00104588,0.0003848672,0.002637224,0.05460523,0.0087868],"genre_scores_gemma":[0.9035788,0.0004225173,0.08199412,0.0007757584,0.0001247099,0.0001896966,0.005537067,0.0008099097,0.00656738],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005598349,"threshold_uncertainty_score":0.01852012,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02492057637310018,"score_gpt":0.2507083064115961,"score_spread":0.2257877300384959,"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."}}