---
title: "Vitou Phy (@vitouphy) | Hashnode"
description: "Software Developer. Read the latest articles by Vitou Phy on Hashnode."
url: "https://hashnode.com/@vitouphy"
---

@vitouphy

# Vitou Phy

@vitouphy · Tokyo, Japan · Joined October 2022

Software Developer

### [A Model that Retrieves Documents When in Doubt](https://demystifyml.co/a-model-that-retrieves-documents-when-in-doubt?utm_source=hashnode&utm_medium=feed)

Summary of Active retrieval augmented generation paper

4 min read · [demystifyml.co](https://demystifyml.co/a-model-that-retrieves-documents-when-in-doubt?utm_source=hashnode&utm_medium=feed)

[Dec 24, 2023](https://hashnode.com/posts/a-model-that-retrieves-documents-when-in-doubt/6588477957379857fc63a6a9)

### [Training MNIST with MLX on Apple Silicon](https://demystifyml.co/training-mnist-with-mlx-on-apple-silicon?utm_source=hashnode&utm_medium=feed)

Classifying digits using Apple M1 GPU

3 min read · [demystifyml.co](https://demystifyml.co/training-mnist-with-mlx-on-apple-silicon?utm_source=hashnode&utm_medium=feed)

[Dec 18, 2023](https://hashnode.com/posts/training-mnist-with-mlx-on-apple-silicon/657ffc50812dc5e15cdc9e05)

### [Augmented SBERT: Data Augmentation Method for Improving Bi-Encoders for Pairwise Sentence Scoring Tasks](https://demystifyml.co/augmented-sbert-data-augmentation-method-for-improving-bi-encoders-for-pairwise-sentence-scoring-tasks?utm_source=hashnode&utm_medium=feed)

Narrowing performance gap with cross-encoder using weakly labeled data

3 min read · [demystifyml.co](https://demystifyml.co/augmented-sbert-data-augmentation-method-for-improving-bi-encoders-for-pairwise-sentence-scoring-tasks?utm_source=hashnode&utm_medium=feed)

[Jul 31, 2023](https://hashnode.com/posts/augmented-sbert-data-augmentation-method-for-improving-bi-encoders-for-pairwise-sentence-scoring-tasks/64c6f98b63af242a13600175)

### [LLaMA: Open and Efficient Foundation Language Models](https://demystifyml.co/llama-open-and-efficient-foundation-language-models?utm_source=hashnode&utm_medium=feed)

Large Language Model for Commercial Purpose

2 min read · [demystifyml.co](https://demystifyml.co/llama-open-and-efficient-foundation-language-models?utm_source=hashnode&utm_medium=feed)

[Jul 25, 2023](https://hashnode.com/posts/llama-open-and-efficient-foundation-language-models/64bf108de301fae450aa673f)

### [DeiT: Data-Efficient Image Transformer](https://demystifyml.co/deit-data-efficient-image-transformer?utm_source=hashnode&utm_medium=feed)

Training data-efficient image transforms & distillation through attention

3 min read · [demystifyml.co](https://demystifyml.co/deit-data-efficient-image-transformer?utm_source=hashnode&utm_medium=feed)

[Jul 10, 2023](https://hashnode.com/posts/deit-data-efficient-image-transformer/64ab4a29a25b102014463961)