Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206460194> ?p ?o ?g. }
- W3206460194 abstract "Prompt learning methods adapt pre-trained language models to downstream applications by using a task-specific prompt together with the input. Most of the current work on prompt learning in text generation relies on a shared dataset-level prompt for all examples in the dataset. We extend this approach and propose a dynamic method, Control Prefixes, which allows for the inclusion of conditional input-dependent information in each prompt. Control Prefixes is at the intersection of prompt learning and controlled generation, empowering the model to have finer-grained control during text generation. The method incorporates attribute-level learnable representations into different layers of a pre-trained transformer, allowing for the generated text to be guided in a particular direction. We provide a systematic evaluation of the technique and apply it to five datasets from the GEM benchmark for natural language generation (NLG). We present state-of-the-art results on several data-to-text datasets, including WebNLG." @default.
- W3206460194 created "2021-10-25" @default.
- W3206460194 creator A5029820246 @default.
- W3206460194 creator A5049521974 @default.
- W3206460194 creator A5062209641 @default.
- W3206460194 date "2021-10-15" @default.
- W3206460194 modified "2023-09-27" @default.
- W3206460194 title "Control Prefixes for Text Generation." @default.
- W3206460194 cites W102708294 @default.
- W3206460194 cites W1507711477 @default.
- W3206460194 cites W2101105183 @default.
- W3206460194 cites W2133512280 @default.
- W3206460194 cites W2149327368 @default.
- W3206460194 cites W2154652894 @default.
- W3206460194 cites W2187089797 @default.
- W3206460194 cites W2250539671 @default.
- W3206460194 cites W2534253848 @default.
- W3206460194 cites W2550821151 @default.
- W3206460194 cites W2557449848 @default.
- W3206460194 cites W2768282280 @default.
- W3206460194 cites W2786660442 @default.
- W3206460194 cites W2796032388 @default.
- W3206460194 cites W2797328513 @default.
- W3206460194 cites W2888482885 @default.
- W3206460194 cites W2888882903 @default.
- W3206460194 cites W2953033958 @default.
- W3206460194 cites W2962815673 @default.
- W3206460194 cites W2963104691 @default.
- W3206460194 cites W2968888988 @default.
- W3206460194 cites W2970352191 @default.
- W3206460194 cites W2973049837 @default.
- W3206460194 cites W2974535386 @default.
- W3206460194 cites W2982399380 @default.
- W3206460194 cites W2990761674 @default.
- W3206460194 cites W2996176596 @default.
- W3206460194 cites W2996287690 @default.
- W3206460194 cites W2997195635 @default.
- W3206460194 cites W3022915354 @default.
- W3206460194 cites W3023237241 @default.
- W3206460194 cites W3034715004 @default.
- W3206460194 cites W3035456141 @default.
- W3206460194 cites W3039127676 @default.
- W3206460194 cites W3042876905 @default.
- W3206460194 cites W3082274269 @default.
- W3206460194 cites W3098824823 @default.
- W3206460194 cites W3099396524 @default.
- W3206460194 cites W3118216348 @default.
- W3206460194 cites W3119438769 @default.
- W3206460194 cites W3125833936 @default.
- W3206460194 cites W3131933120 @default.
- W3206460194 cites W3139080614 @default.
- W3206460194 cites W3152956381 @default.
- W3206460194 cites W3153427360 @default.
- W3206460194 cites W3154592693 @default.
- W3206460194 cites W3168867926 @default.
- W3206460194 cites W3170083118 @default.
- W3206460194 cites W3177174258 @default.
- W3206460194 cites W3037551068 @default.
- W3206460194 cites W3135863680 @default.
- W3206460194 hasPublicationYear "2021" @default.
- W3206460194 type Work @default.
- W3206460194 sameAs 3206460194 @default.
- W3206460194 citedByCount "0" @default.
- W3206460194 crossrefType "posted-content" @default.
- W3206460194 hasAuthorship W3206460194A5029820246 @default.
- W3206460194 hasAuthorship W3206460194A5049521974 @default.
- W3206460194 hasAuthorship W3206460194A5062209641 @default.
- W3206460194 hasConcept C119857082 @default.
- W3206460194 hasConcept C121332964 @default.
- W3206460194 hasConcept C13280743 @default.
- W3206460194 hasConcept C137293760 @default.
- W3206460194 hasConcept C138885662 @default.
- W3206460194 hasConcept C141603448 @default.
- W3206460194 hasConcept C152565575 @default.
- W3206460194 hasConcept C154945302 @default.
- W3206460194 hasConcept C165801399 @default.
- W3206460194 hasConcept C185798385 @default.
- W3206460194 hasConcept C204321447 @default.
- W3206460194 hasConcept C205649164 @default.
- W3206460194 hasConcept C2775924081 @default.
- W3206460194 hasConcept C2985684807 @default.
- W3206460194 hasConcept C41008148 @default.
- W3206460194 hasConcept C41895202 @default.
- W3206460194 hasConcept C62520636 @default.
- W3206460194 hasConcept C66322947 @default.
- W3206460194 hasConceptScore W3206460194C119857082 @default.
- W3206460194 hasConceptScore W3206460194C121332964 @default.
- W3206460194 hasConceptScore W3206460194C13280743 @default.
- W3206460194 hasConceptScore W3206460194C137293760 @default.
- W3206460194 hasConceptScore W3206460194C138885662 @default.
- W3206460194 hasConceptScore W3206460194C141603448 @default.
- W3206460194 hasConceptScore W3206460194C152565575 @default.
- W3206460194 hasConceptScore W3206460194C154945302 @default.
- W3206460194 hasConceptScore W3206460194C165801399 @default.
- W3206460194 hasConceptScore W3206460194C185798385 @default.
- W3206460194 hasConceptScore W3206460194C204321447 @default.
- W3206460194 hasConceptScore W3206460194C205649164 @default.
- W3206460194 hasConceptScore W3206460194C2775924081 @default.
- W3206460194 hasConceptScore W3206460194C2985684807 @default.
- W3206460194 hasConceptScore W3206460194C41008148 @default.