Retrieval-Augmented Generation (RAG) using Large Language Models (LLMs)

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F24%3A43973058" target="_blank" >RIV/49777513:23520/24:43973058 - isvavai.cz</a>

Výsledek na webu

<a href="https://svk.fav.zcu.cz/download/proceedings_svk_2024.pdf" target="_blank" >https://svk.fav.zcu.cz/download/proceedings_svk_2024.pdf</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Retrieval-Augmented Generation (RAG) using Large Language Models (LLMs)

Popis výsledku v původním jazyce

The convergence of Retrieval-augmented generation (RAG) methodologies with the ro- bust computational prowess of Large Language Models (LLMs) heralds a new era in natural language processing, promising unprecedented levels of accuracy and contextual relevance in text generation tasks. Pre-trained large language models, also referred to as foundation models, typically lack the ability to learn incrementally, may exhibit hallucinations, and can inadvertently expose pri- vate data from their training corpus. Addressing these shortcomings has sparked increasing interest in retrieval-augmented generation methods. RAG enhances the predictive capabilities of large language models by integrating an external datastore during inference. This approach enriches prompts with a blend of context, historical data, and pertinent knowledge, resulting in RAG LLMs.

Název v anglickém jazyce

Retrieval-Augmented Generation (RAG) using Large Language Models (LLMs)

Popis výsledku anglicky

The convergence of Retrieval-augmented generation (RAG) methodologies with the ro- bust computational prowess of Large Language Models (LLMs) heralds a new era in natural language processing, promising unprecedented levels of accuracy and contextual relevance in text generation tasks. Pre-trained large language models, also referred to as foundation models, typically lack the ability to learn incrementally, may exhibit hallucinations, and can inadvertently expose pri- vate data from their training corpus. Addressing these shortcomings has sparked increasing interest in retrieval-augmented generation methods. RAG enhances the predictive capabilities of large language models by integrating an external datastore during inference. This approach enriches prompts with a blend of context, historical data, and pertinent knowledge, resulting in RAG LLMs.

Druh

O - Ostatní výsledky

CEP obor

—

OECD FORD obor

20205 - Automation and control systems

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů