graph-of-thoughts and Algorithm-Of-Thoughts
These are competing approaches to structured LLM reasoning that both augment chain-of-thought prompting with graph-based exploration of solution spaces, but Graph of Thoughts uses explicit graph construction while Algorithm of Thoughts uses tree-based thought branching.
About graph-of-thoughts
spcl/graph-of-thoughts
Official Implementation of "Graph of Thoughts: Solving Elaborate Problems with Large Language Models"
This framework helps AI/ML engineers and researchers design more effective large language model (LLM) workflows for complex tasks. It takes a problem definition and an LLM, then orchestrates the LLM's 'thought process' through a series of operations, like generating ideas or scoring options, to arrive at a solution. The output is a structured graph detailing the LLM's problem-solving steps.
About Algorithm-Of-Thoughts
kyegomez/Algorithm-Of-Thoughts
My implementation of "Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models"
This project helps AI developers make large language models (LLMs) better at complex problem-solving and reasoning tasks. By providing a structured way to explore different ideas and backtrack from dead ends, it takes a problem statement and helps the LLM arrive at a more accurate and robust solution. It's designed for AI practitioners and researchers who build and refine LLMs.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work