Ahmet's picture

Ahmet

atasoglu

·

atasoglu

AI & ML interests

NLP, LLMs.

Recent Activity

New activity 2 days ago

Metin/Gemma-2-9b-it-TR-DPO-V1

updated a model 2 days ago

Metin/Gemma-2-9b-it-TR-DPO-V1

upvoted a paper 3 days ago

Organizations

atasoglu's activity

upvoted a paper 3 days ago

LLaVA-o1: Let Vision Language Models Reason Step-by-Step

Paper • 2411.10440 • Published 6 days ago • 87

upvoted a collection 6 days ago

Nov 15 Releases 🍂

15 items • Updated 6 days ago • 6

upvoted 2 collections 21 days ago

MobileLLM

Optimizing Sub-billion Parameter Language Models for On-Device Use Cases (ICML 2024) https://arxiv.org/abs/2402.14905 • 8 items • Updated 14 days ago • 95

SmolLM2

State-of-the-art compact LLMs for on-device applications: 1.7B, 360M, 135M • 10 items • Updated about 1 hour ago • 172

upvoted an article 21 days ago

Article

Model2Vec: Distill a Small Fast Model from any Sentence Transformer

By

•

Oct 14

• 55

upvoted 2 collections 23 days ago

LayerSkip

Models continually pretrained using LayerSkip - https://arxiv.org/abs/2404.16710 • 8 items • Updated about 1 hour ago • 43

MIT Talk 31/10 Papers

14 items • Updated 24 days ago • 29

upvoted a collection 27 days ago

October 25 Releases

19 items • Updated 27 days ago • 7

upvoted a paper 29 days ago

Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Prediction

Paper • 2409.18124 • Published Sep 26 • 31

upvoted a collection 29 days ago

LOTUS 🪷

8 items • Updated 29 days ago • 5

upvoted 2 articles about 1 month ago

Article

🇮🇹🇯🇵🇧🇷 Generating multilingual instruction datasets with Magpie 🐦‍⬛

By

•

Oct 21

• 18

Article

Fixing Gradient Accumulation

Oct 16

• 41

upvoted a collection about 1 month ago

Llama-3.1-Nemotron-70B

SOTA models on Arena Hard and RewardBench as of 1 Oct 2024. • 6 items • Updated Oct 15 • 139

upvoted a paper about 1 month ago

EuroLLM: Multilingual Language Models for Europe

Paper • 2409.16235 • Published Sep 24 • 24

upvoted a paper about 2 months ago

Were RNNs All We Needed?

Paper • 2410.01201 • Published Oct 2 • 47

upvoted 3 collections about 2 months ago

Qwen 2.5

27 items • Updated about 11 hours ago • 3

Vision-TR-Open-Datasets

This collection contains Turkish multimodal datasets that are suitable for the task of Image-Text-to-Text. • 3 items • Updated Sep 8 • 1

Computer Vision Backbones 🧩

Collection of useful computer vision backbones to fine-tune. It also includes large image classification models, that can be used as backbone. • 22 items • Updated Sep 19, 2023 • 19

upvoted a paper about 2 months ago

NVLM: Open Frontier-Class Multimodal LLMs

Paper • 2409.11402 • Published Sep 17 • 72

upvoted a collection about 2 months ago

Core ML Stable Diffusion

16 items • Updated Oct 4 • 14