ariz565 / customized-llm

A comprehensive list about how to customize your own LLM

Customized LLM

Make LLM a personal assistant.

Papers

• (2023-05) PALR: Personalization Aware LLMs for Recommendation arxiv

• (2023-05) Augmented Large Language Models with Parametric Knowledge Guiding arxiv

• (2023-05) Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction arxiv

• (2023-05) PersonaLLM: Investigating the Ability of GPT-3.5 to Express Personality Traits and Gender Differences arxiv

• (2023-05) Can LLM Already Serve as A Database Interface? A BIg Bench for Large-Scale Database Grounded Text-to-SQLs arxiv

  external knowledge for Text-to-SQLs tasks. Simply retrieve and then concate.

• (2023-05) Privacy-Preserving Prompt Tuning for Large Language Model Services arxiv

  T5/BERT

• (2023-04) Unstructured and structured data: Can we have the best of both worlds with large language models? arxiv

• (2023-04) LaMP: When Large Language Models Meet Personalization arxiv

• (2023-04) Emergent and Predictable Memorization in Large Language Models arxiv

• (2022-12) Personalizing Large Language Models link

• (2021-ICTSD) Towards User-Centric Text-to-Text Generation: A Survey link

• (2020-LREC) Evaluating Approaches to Personalizing Language Models LREC

• (2018-ACL) Personalized Language Model for Query Auto-Completion ACL-Short

Datasets

• LaMP: When Large Language Models Meet Personalization link

  Citation Identification, News Categorization, Product Rating, News Headline Generation, Scholarly Title Generation, Email Subject Generation,Tweet Paraphrasing

• RecBole-PJF: Person Job Fitting link

• PersonaChat - A chit-chat dataset where paired Turkers are given assigned personas and chat to try to get to know each other. link

• Blog Authorship Corpus - this dataset contains text from blogs written on or before 2004, with each blog being the work of a single user. link

• Project Gutenberg - Project Gutenberg is a library of over 70,000 free eBooks link

• Song Lyrics - It contains song lyrics of modern singers (British and American English), as well as some English poetry. In general, there are 49 authors. link

Resources

• LlamaIndex (GPT Index) is a project that provides a central interface to connect your LLM’s with external data.

• [NVIDIA] A Step-by-step Guide to Building Large Custom Language Models link

• NeMo is a cloud service that kick-starts the journey to hyper-personalized enterprise AI applications

• LangChain is a framework for developing applications powered by language models.

• privateGPT Ask questions to your documents without an internet connection, using the power of LLMs.

Opinions

• The Dual LLM pattern for building AI assistants that can resist prompt injection link
• User Personalization, the Next Frontier in LLMs link