👉🏼 On-device query intent prediction with lightweight LLMs to support ubiquitous conversations 🤓 Mateusz Dubiel 👇🏻 https://lnkd.in/ekpGf3ct 🔍 Focus on data insights: - Lightweight pre-trained LLMs fine-tuned for on-device query intent prediction - Transfer learning techniques applied to enhance flexibility and scalability of conversational agents - Privacy-preserving scenarios enabled by on-device deployment of LLMs 💡 Main outcomes and implications: - Improved performance of RoBERTa and XLNet in predicting user query intent - Comparable performance of fine-tuned LLMs with ChatGPT - Highlighting the balance between LLM performance, memory footprint, and privacy considerations 📚 Field significance: - Advancing the field of conversational agents towards more flexible and scalable models - Addressing privacy concerns by enabling on-device deployment of LLMs - Providing insights for researchers and practitioners on leveraging LLMs for personalized and privacy-preserving conversational experiences 🗄️: #ConversationalAgents #LLMs #OnDeviceDeployment #PrivacyPreservation #QueryIntentPrediction
Nick Tarazona, MD’s Post
More Relevant Posts
-
Building AI products @ E.ON Germany | Azure AI Engineer, MLOPS, LLM, Distributed Computing
LLM Blending is All You Need: Cheaper, Better Alternative to Trillion-Parameters LLM Blending ensembles, or randomly selecting from just 3 different 6B/13B parameter models at each generation of a chat response, conditioning on chat history, outperformed 175B+ parameter ChatGPT. 🤯 Not only do Blended chat AIs have higher user retention, indicating that they are more engaging/useful, but their inference costs are equivalent to a single 6B/13B system. State-of-the-art chat AIs follow a three-stage pipeline involving pre-trained language models, reward models, and fine-tuning We know that optimizing a model for one metric, like reasoning, or mathematics usually reduces its performance on other metrics, for example, reading comprehension. Ensembling multiple chat AIs can lead to a system with overall better characteristics based on Bayesian statistical principles Smaller models are king 👑 Paper: https://lnkd.in/d69vFJxC Follow for more content like this. #llm #esemble #blending #chatgpt #openai #microsoft #eon
-
-
STUDYING ARTIFICIAL INTELLIGENCE AND MACHINE, Fellow at NxtWave’s CCBP 4.0 Academy | Knows Python, Front End Development | Looking for internships.
Just finished Chat with Your Data Using ChatGPT! Check it out: https://lnkd.in/eUVvdRqK #chatbotdevelopment #largelanguagemodels #naturallanguageprocessing
-
Just finished Chat with Your Data Using ChatGPT! Check it out: https://lnkd.in/g9B92BgT #chatbotdevelopment #largelanguagemodels #naturallanguageprocessing
Certificate of Completion
linkedin.com
-
Just finished Chat with Your Data Using ChatGPT! Check it out: https://lnkd.in/dwDF5X3K #chatbotdevelopment #largelanguagemodels #naturallanguageprocessing
Certificate of Completion
linkedin.com
-
undergraduate computer science [email protected] || Fresher || mobile app development || databse
I want to talk about my assignment in which Architecture level difference between GOOGLE AI gemini and chatgpt , advantages and disadvantages, future uses will be discussed.
-
-
Just finished Chat with Your Data Using ChatGPT! Check it out: https://lnkd.in/gbrzhdB5 #chatbotdevelopment #largelanguagemodels #naturallanguageprocessing
Certificate of Completion
linkedin.com
-
Student of Computer Applications| Artificial Intelligence, Data Science, HTML, CSS, JavaScript
Just finished Chat with Your Data Using ChatGPT! Check it out: https://lnkd.in/dedQPX7V #chatbotdevelopment #largelanguagemodels #naturallanguageprocessing
Certificate of Completion
linkedin.com
-
Solution Architect/Technology Architect/Innovator/Thought leader/Author/Technology writer
Gen AI-powered LLM for Noun Detection in Search Queries As part of my ongoing project to develop a Gen AI-enabled photo album system, I have created a Node.js-based front-end that replicates the interface of ChatGPT. The system uses a Python-powered API for noun detection in user search queries, allowing users to easily search and retrieve images from Azure storage by identifying relevant nouns in the query. While the project currently includes face detection, the accuracy is still being improved, and the next focus will be on providing an automated metadata correction step.
-
Performance Test Engineer with 9 Years of Experience | Expert with Apache JMeter | Tech tools - Blazemeter, Micro Focus LoadRunner, Gatling, New Relic, K6, StresStimulus, AppDynamics, Dynatrace | Cloud - Azure, AWS
Just finished Chat with Your Data Using ChatGPT! Check it out: https://lnkd.in/gyNT4feV #chatbotdevelopment #largelanguagemodels #naturallanguageprocessing
Certificate of Completion
linkedin.com
-
Polyvalent Surgeon, CV updated. DRAFT PROFILE. Taking a few days off, redefining my priorities and generating a new profile.
Just finished Chat with Your Data Using ChatGPT! Check it out: https://lnkd.in/dFYfGhSA #chatbotdevelopment #largelanguagemodels #naturallanguageprocessing
Certificate of Completion
linkedin.com
