How to build Google-like AI-powered search using Python in 15 minutes? (A thread)
LLMS
-
Chat with your documents: Filechat.io offers unlimited AI interaction with PDFs
By
–
Use ChatGPT on your own data (without limits)https://t.co/g8gwEboOZ9
— AI Breakfast (@AiBreakfast) 6 février 2023
Upload a PDF (any size), have a conversational dialogue with your documents:
-Research Papers
-Books
-Newspapers
-Legal Docs
Pulls citations + unlimited use plan available
Semantic search is the future ↓ pic.twitter.com/VDXJe9ZpnlUse ChatGPT on your own data (without limits) http://
Filechat.io Upload a PDF (any size), have a conversational dialogue with your documents: -Research Papers
-Books
-Newspapers
-Legal Docs Pulls citations + unlimited use plan available Semantic search is the future ↓ -
Elf-Attention Models and the ORC Challenge
By
–
Can models based on the Elf-Attention mechanism tackle the ORC challenge?
-
LLM Focus May Delay Path to AGI Development
By
–
I have often made same argument; agree with @ylecun that excessive focus on LLMs may delay HLAI/AGI
-
Chain-of-Thought Prompting Enhances Language Model Reliability
By
–
Fascinating work combining chain-of-thought prompting and tool use to improve the reliability of language models.
-
LLMs Are English-to-English Machine Translation Systems
By
–
Lost on most people: every time you used http://
translate.google.com for past 20 years, an LLM was used. The main difference now: we’re translating English prompts to chat-inspired predictions also in English. LLMs are basically English-to-English machine translation -
Transfer Learning in NLP Workshop Research Presentation
By
–
Work by Rafal Kocielnik, @SaraKangaslahti @shrimai_ @rmichaelalvarez @caltech @nvidia Appeared at #Neurips workshop on transfer learning in NLP
-
Active Learning and Transfer Learning in Few-Shot LLM Settings
By
–
How can we mix active and transfer learning in few-shot learning setting with pre-trained LLMs? We show that you need to label only few samples in-domain and leverage on transfer learning from out-domain along with ingrained knowledge in pretrained models
-
REPLUG: Retrieval-Augmented Black-Box Language Models
By
–
REPLUG: Retrieval-Augmented Black-Box Language Models Shi et al.: https://
arxiv.org/abs/2301.12652 #Artificialintelligence #DeepLearning #MachineLearning -
Historical AI Agreements and Paradigm Shift Debates
By
–
Amazing how many people believe this, even when I have written extensively (even this week!) about agreements going back several years. Eg see here 2017 slide https://
garymarcus.substack.com/p/some-things-
garymarcus-might-say
… and see long essay on 2022 https://
garymarcus.substack.com/p/does-ai-real
ly-need-a-paradigm-shift
…