Cobus Greyling

Time-Aware Adaptive RAG (TA-ARE)

With the advent of Large Language Models (LLMs) there was the concept of emergent capabilities. The premise or...

Proxy Fine-Tuning LLMs

Proxy fine-tuning achieves the results of directly tuning a LLM, but by accessing only its prediction…

But first let’s...

Fine-Tuning or RAG?

Comparing different LLM knowledge injection methods…Fine-Tuning or RAG? The short answer is, it depends…

There are a...

T-RAG = RAG + Fine-Tuning + Entity Detection

The T-RAG approach is premised on combining RAG architecture with an open-source fine-tuned LLM and an entities tree ...

LLM Drift, Prompt Drift & Cascading

Prompt Chaining can be performed manually or automatically; manual entails crafting chains by hand, via a GUI chain...

The Case For Small Language Models

Considering Conversational AI implementations in general, like chatbots and voicebots, making use of a Large...

Beyond Chain-of-Thought LLM Reasoning

This approach can be implemented on a prompt level and does not require any dedicated frameworks or pre-processing.

A...

Comparing Human, LLM & LLM-RAG Responses

A recent study, focusing on the healthcare & preoperative medicine compared expert human feedback with LLM...

A Benchmark for Verifying Chain-Of-Thought

A Chain-of-Thought is only as strong as its weakest link; a recent study from Google Research created a benchmark...

Seven RAG Engineering Failure Points

Retrieval-Augmented Generation (RAG) systems remains a compelling solution to the challenge of relevant up-to-date...