Computer implemented method for obtaining an output response to a natural language input prompt, the output response being generated by a language model selected from a group of language models, comprising the steps: - receiving the input prompt; 10 - determining a complexity score of the input prompt; - determining a time-criticality score of the input prompt; - for each language model, computing a quality of service score related to the combination of the language model 15 and the input prompt, on the basis of the complexity score of the input prompt, the time-criticality score of the input prompt, a stored response quality function of the language model and a stored latency function of the language model; 20 - selecting a selected language model on the basis of the computed quality service; - forwarding the input prompt to the selected language model; - receiving the output response generated by the selected 25 language model on the basis of the input prompt.

COMPUTER IMPLEMENTED METHOD FOR OBTAINING AN OUTPUT RESPONSE TO A NATURAL LANGUAGE INPUT PROMPT

Roveri Manuel;SHRESHTH TULI;CASALE Giuliano;
2024-01-01

Abstract

Computer implemented method for obtaining an output response to a natural language input prompt, the output response being generated by a language model selected from a group of language models, comprising the steps: - receiving the input prompt; 10 - determining a complexity score of the input prompt; - determining a time-criticality score of the input prompt; - for each language model, computing a quality of service score related to the combination of the language model 15 and the input prompt, on the basis of the complexity score of the input prompt, the time-criticality score of the input prompt, a stored response quality function of the language model and a stored latency function of the language model; 20 - selecting a selected language model on the basis of the computed quality service; - forwarding the input prompt to the selected language model; - receiving the output response generated by the selected 25 language model on the basis of the input prompt.
2024
File in questo prodotto:
File Dimensione Formato  
DEIB.25.0327.C_EP__English_specification.pdf

Accesso riservato

Descrizione: english text
: Altro materiale allegato
Dimensione 94.89 kB
Formato Adobe PDF
94.89 kB Adobe PDF   Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11311/1290644
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact