Benjamin Paine's picture

Benjamin Paine PRO

benjamin-paine

·

https://www.benjaminpaine.com

AI & ML interests

A software engineer with an AI habit

Recent Activity

reacted to frimelle's post with ❤️ 3 days ago

What’s in a name? More than you might think, especially for AI. Whenever I introduce myself, people often start speaking French to me, even though my French is très basic. It turns out that AI systems do something similar: Large language models infer cultural identity from names, shaping their responses based on presumed backgrounds. But is this helpful personalization or a reinforcement of stereotypes? In our latest paper, we explored this question by testing DeepSeek, Llama, Aya, Mistral-Nemo, and GPT-4o-mini on how they associate names with cultural identities. We analysed 900 names from 30 cultures and found strong assumptions baked into AI responses: some cultures were overrepresented, while others barely registered. For example, a name like "Jun" often triggered Japan-related responses, while "Carlos" was linked primarily to Mexico, even though these names exist in multiple countries. Meanwhile, names from places like Ireland led to more generic answers, suggesting weaker associations in the training data. This has real implications for AI fairness: How should AI systems personalize without stereotyping? Should they adapt at all based on a name? Work with some of my favourite researchers: @sidicity Arnav Arora and @IAugenstein Read the full paper here: https://huggingface.co/papers/2502.11995

liked a Space 3 days ago

nanotron/ultrascale-playbook

new activity 3 days ago

benjamin-paine/Lumina-Image-2.0:Upgradio

View all activity

Organizations

benjamin-paine's activity

upvoted an article 23 days ago

Article

The AI tools for Art Newsletter - Issue 1

24 days ago

• 64

upvoted an article 26 days ago

Article

Welcome to Inference Providers on the Hub 🔥

27 days ago

• 384

upvoted a collection 4 months ago

Hey Buddy!

Hey Buddy! is an open-source library for training wake word models (a.k.a audio keyword spotters) and deploying them to the browser. • 3 items • Updated Oct 25, 2024 • 3

upvoted 3 collections 5 months ago

Free Music Archive

ISMIR's 2017 FMA Dataset, Optimized for 🤗 Datasets / 🥐 Croissant, with Clear Licensing • 4 items • Updated Sep 13, 2024 • 4

Audio Augmentation Pack (16khz)

Downsampled Audio Datasets with Open Licenses, Perfect for Data Augmentation • 3 items • Updated Sep 13, 2024 • 2

ImageNet 1K

ILSVRC's 2012 ImageNet 1K Subset (1.43M images,) Optimized for 🤗 Datasets / 🥐 Croissant, with Pre-Cropped and Scaled Versions • 5 items • Updated Sep 15, 2024 • 2

upvoted a collection 6 months ago

Resources for Sound Processing

647 items • Updated about 20 hours ago • 9

upvoted a paper 6 months ago

Transformer Explainer: Interactive Learning of Text-Generative Models

Paper • 2408.04619 • Published Aug 8, 2024 • 159