---
title: "Guillaume Guy (@guillaume-guy) | Hashnode"
description: "AI, ML, Deep Learning. Read the latest articles by Guillaume Guy on Hashnode."
url: "https://hashnode.com/@guillaume-guy"
---

@guillaume-guy

# Guillaume Guy

@guillaume-guy · Joined April 2025

AI, ML, Deep Learning

### [What is CLIP and why does it matter?](https://www.guillaume.nyc/what-is-clip-and-why-does-it-matter?utm_source=hashnode&utm_medium=feed)

Paper: Learning Transferable Visual Models From Natural Language Supervision, Radford et al. (2021) Introduction As we discussed in previous posts, Contrastive Learning isn't new. For example, FAIR's Moco was published in 2019, along with many other ...

3 min read · [guillaume.nyc](https://www.guillaume.nyc/what-is-clip-and-why-does-it-matter?utm_source=hashnode&utm_medium=feed)

[May 12, 2025](https://hashnode.com/posts/what-is-clip-and-why-does-it-matter/6822054e62f6347293451e38)

### [How did LLMs gain Vision?](https://www.guillaume.nyc/how-did-llms-gain-vision?utm_source=hashnode&utm_medium=feed)

Introduction LLMs were created as text-only chat bots as an artifact of their training paradigm: they learn to predict the next token (~word). Photos, videos and other richer media don’t have words and therefore were naturally excluded from the train...

3 min read · [guillaume.nyc](https://www.guillaume.nyc/how-did-llms-gain-vision?utm_source=hashnode&utm_medium=feed)

[May 5, 2025](https://hashnode.com/posts/how-did-llms-gain-vision/6818d50d904270e1f4976860)

### [The state of Self-Supervision for Vision](https://www.guillaume.nyc/the-state-of-self-supervision-for-vision?utm_source=hashnode&utm_medium=feed)

Introduction To perform vision tasks effectively, it's important to have a strong, general-purpose vision backbone. This allows you to handle many vision tasks, such as: Image-to-image similarity: For comparison or retrieval. Vision adapters: Add a...

7 min read · [guillaume.nyc](https://www.guillaume.nyc/the-state-of-self-supervision-for-vision?utm_source=hashnode&utm_medium=feed)

[May 1, 2025](https://hashnode.com/posts/the-state-of-self-supervision-for-vision/6813958c283a51717d4c52ec)

### [Don't use raw embeddings](https://www.guillaume.nyc/dont-use-raw-embeddings?utm_source=hashnode&utm_medium=feed)

Introduction: With the rise of Transformers, embeddings are now widely used: As representations of images or texts that can be used by other models or in a zero-shot manner As a basic building block for Vector Search in LLM RAG and image search H...

3 min read · [guillaume.nyc](https://www.guillaume.nyc/dont-use-raw-embeddings?utm_source=hashnode&utm_medium=feed)

[Apr 15, 2025](https://hashnode.com/posts/dont-use-raw-embeddings/67fec6c1ff98a3ee053f6a6a)

### [Focus: Shapley value](https://www.guillaume.nyc/focus-shapley-value?utm_source=hashnode&utm_medium=feed)

Understanding Shapley Value

2 min read · [guillaume.nyc](https://www.guillaume.nyc/focus-shapley-value?utm_source=hashnode&utm_medium=feed)

[Apr 15, 2025](https://hashnode.com/posts/focus-shapley-value/67fe87c76684650a45af7c40)
