Amirhossein Layegh

Amirhossein Layegh

PhD Student

Blog posts

2025

From Local to Global Sensemaking: First Impressions of Microsoft GraphRAG (MS GraphRAG)

7 minute read

Published:

TL;DR GraphRAG replaces vector search with a lightweight knowledge-graph index and a map-reduce summarization step. The result: LLMs can tackle global questions such as “What themes span this entire corpus?” while remaining fast and token-efficient. In head-to-head tests against GPT-4-powered vector RAG, GraphRAG won 72-83 % of comparisons on answer comprehensiveness and 62-82 % on diversity, while using up to 97 % fewer context tokens for some query modes.

2024

Stream of Search (SoS): Learning to Search in Language

2 minute read

Published:

In the landscape of AI, the ability of LLMs to reason, plan and solve complex problems has become increasingly significant. This novel paper titled “Stream of Search (SoS): Learning to Search in Language” delves into this crucial area, proposing an approach that enhances the reasoning capabilities of LLMs through the concept of search.

Reflection Tuning

4 minute read

Published:

In this post, we will review one of the most recent and effective approaches for improving the quality of instruction-tuning data, known as reflection-tuning.

Hands on Transformers

10 minute read

Published:

In this post we will be looking at the Transformer architecture in detail. You can also find the code for this post on my GitHub Repository.

2023

Prompt Engineering Guide

20 minute read

Published:

Prompt Engineering

Prompt engineering is a discipline focused on crafting efficient prompts for large language models (LLMs). These skills aid in comprehending LLM capabilities and limitations. Researchers apply prompt engineering to enhance LLM performance in various tasks, from QA to arithmetic reasoning. Developers use it to create compelling interfaces for LLMs and other tools.

NLP Paper Club Week1 (Sep 25 - Oct 1)

18 minute read

Published:

In this blog post I tried to introduce top papers in ML (particularly in NLP) and give a brief overview on them.