Article
Fairness Pruning: Finding Demographic Bias Neurons in LLMs (And What Happens When You Remove Them)
oopere
• • 1Pere Martra
oopere
AI & ML interests
AI Research Engineer | Structured Pruning & LLM Efficiency | Fairness-Aware Optimization | Author “Rearchitecting LLMs” @Manning | 2K GitHub ⭐
Recent Activity
liked a model about 5 hours ago
oracomputing/LlamaOra-6.4B-Instruct updated a dataset 5 days ago
oopere/clinical-ner-qdora updated a model 5 days ago
oopere/gemma-3-270m-14L-distilledOrganizations
upvoted an article 5 days ago
upvoted a collection about 2 months ago
upvoted a paper 5 months ago
upvoted an article 7 months ago
Article
Making LLMs Smaller Without Breaking Them: A GLU-Aware Pruning Approach
oopere
• • 20 upvoted 2
articles 11 months ago
Article
Good answers are not necessarily factual answers: an analysis of hallucination in leading LLMs
davidberenstein1957
• • 42 Article
LLMs recognise bias but also reproduce harmful stereotypes: an analysis of bias in leading LLMs
davidberenstein1957
• • 16 upvoted an article 12 months ago
Article
Welcome Gemma 3: Google's all new multimodal, multilingual, long context open LLM
- +2
ariG23498, merve, pcuenq, reach-vb
• • 497 upvoted an article about 1 year ago
Article
Training a Gemma 2 2B-IT for Reasoning with GRPO
lmassaron
• • 6 upvoted an article over 1 year ago
Article
Open-R1: a fully open reproduction of DeepSeek-R1
- +1
eliebak, lvwerra, lewtun
• • 889 upvoted a paper over 1 year ago
upvoted 2
articles almost 2 years ago
Article
Preference Tuning LLMs with Direct Preference Optimization Methods
- +3
kashif, edbeeching, lewtun, lvwerra, osanseviero
• • 83 Article
How NuminaMath Won the 1st AIMO Progress Prize
- +6
yfleureau, liyongsea, edbeeching, lewtun, benlipkin, romansoletskyi, vwxyzjn, kashif
• • 128 upvoted a collection about 2 years ago
upvoted 2
articles about 2 years ago
Article
🪆 Introduction to Matryoshka Embedding Models
- +1
tomaarsen, Xenova, osanseviero
• • 210 Article
Making LLMs even more accessible with bitsandbytes, 4-bit quantization and QLoRA
- +3
ybelkada, timdettmers, artidoro, sgugger, smangrul
• • 180