Automatic Metadata Extraction @RLanceMartin has done a lot of work showing how having the correct metadata attached to documents can be valuable, but there's still the question: how do you get that metadata? Use an LLM to extract it of course! Now live in JS and Python
CODE
-
LangChain Integrates Context AI for Chat Analytics
By
–
Chat Application Analytics Introducing an integration with @getcontextai which allows you to receive user analytics with a one line plugin Blog: https://
blog.langchain.dev/langchain-x-co
ntext-building-better-chat-products-with-user-analytics/
… Integration: https://
integrations.langchain.com/callbacks?inte
gration_name=ContextCallbackHandler
… -
Release of curated-transformers: Pure PyTorch library for LLMs
By
–
Excited to release this! A pure @PyTorch library for LLMs and other transformers. It strikes a different trade-off from @huggingface Transformers. curated-transformers won't be up-to-the-minute with new architectures, added verbatim. Instead the implementations are factored out https://
x.com/spacy_io/statu
/spacy_io/status/1679480083801866241
… -
Chrome Extension for Web Page Summarization Using LLMs
By
–
Large language models enable new classes of applications to enrich the web browsing experience. A key use case is summarization. This tutorial demonstrates how to create a Google Chrome extension that summarizes a web page using the Summarize endpoint. https://
short.cohere.ai/EQ9HPS -
Gzip and kNN Outperform Transformers for Text Classification
By
–
Gzip + kNN beats transformers on text classification. (Gzip as in good old zip file compression)
-
DNAGPT: Generalized Pretrained Tool for DNA Sequence Analysis
By
–
DNAGPT: A Generalized Pretrained Tool for Multiple DNA Sequence Analysis Tasks paper page: https://
huggingface.co/papers/2307.05
628
… The success of the GPT series proves that GPT can extract general information from sequences, thereby benefiting all downstream tasks. This motivates us to use -
LightGBM Early Stopping Callback for sklearn API
By
–
We need to have early stopping as a callback in the sklearn api. es = lgb.early_stopping(stopping_rounds=200, verbose=False) http://
lgb.fit(train_X, train_y, eval_set=(val_X, val_y), callbacks=[es]) -
Code Interpreter Achieves Near GPT-4 Performance Levels
By
–
Code Interpreter performs close to GPT 4.
-
Scikit-Learn Adds PyTorch GPU Support via Array Dispatch
By
–
wow didn't know this was happening. this is huge!
scikit added support for pytorch and GPUs via array dispatch -
Universal Dependency Annotation Wins 2023 ACL Test of Time Award
By
–
Congratulations to the authors of "Universal dependency annotation for multilingual parsing", which won one of this year's 2023 ACL Test of Time Awards for most influential papers released 10 years ago. Check it out at https://
aclanthology.org/P13-2017.pdf