Occiglot

occiglot's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Articles

stefan-it 

authored a paper about 10 hours ago

SindBERT, the Sailor: Charting the Seas of Turkish NLP

Paper • 2510.21364 • Published 3 days ago • 1

stefan-it 

authored a paper 11 days ago

The German Commons - 154 Billion Tokens of Openly Licensed Text for German Language Models

Paper • 2510.13996 • Published 12 days ago • 6

mbrack 

authored a paper 13 days ago

UniFusion: Vision-Language Model as Unified Encoder in Image Generation

Paper • 2510.12789 • Published 13 days ago • 16

BramVanroy 

posted an update 17 days ago

view post

Post

227

What are currently the best multilingual models with at most 72B parameters? Are Llama 3.3 70B and Qwen 2.5 72B still king?

See translation

• 
• 1 reply

·

👀

1

1

+

Reply

stefan-it 

authored a paper about 2 months ago

Llama-GENBA-10B: A Trilingual Large Language Model for German, English and Bavarian

Paper • 2509.05668 • Published Sep 6 • 5

BramVanroy 

posted an update 2 months ago

view post

Post

715

Thanks to popular request, I've just added two subsets to the CommonCrawl-Creative Commons Corpus (C5; BramVanroy/CommonCrawl-CreativeCommons) so that you do not have to do filtering manually

- C5f ( BramVanroy/CommonCrawl-CreativeCommons-fine): only retains high-quality samples that are also present in FineWeb or FineWeb-2;
- C5r (https://huggingface.co/datasets/BramVanroy/CommonCrawl-CreativeCommons-recommended): additional strict filtering that removes samples with license disagreement, non-commercial licenses, and Wikipedia samples. The latter because you should probably get those from a more reliable source that provides better parsed content.

It goes without saying that these filters lead to a massive reduction in quantity. Doc and token counts are given on the dataset pages.

See translation

👍

3

3

+

Reply

s-conia 

authored a paper 4 months ago

Right Answer, Wrong Score: Uncovering the Inconsistencies of LLM Evaluation in Multiple-Choice Question Answering

Paper • 2503.14996 • Published Mar 19 • 3

mbrack 

authored a paper 4 months ago

How to Train your Text-to-Image Model: Evaluating Design Choices for Synthetic Training Captions

Paper • 2506.16679 • Published Jun 20 • 1

eliaswendt 

authored a paper 5 months ago

Judging Quality Across Languages: A Multilingual Approach to Pretraining Data Filtering with Language Models

Paper • 2505.22232 • Published May 28 • 18

mkrausio 

authored 2 papers 5 months ago

Right on Time: Revising Time Series Models by Constraining their Explanations

Paper • 2402.12921 • Published Feb 20, 2024

Judging Quality Across Languages: A Multilingual Approach to Pretraining Data Filtering with Language Models

Paper • 2505.22232 • Published May 28 • 18

mbrack 

authored 9 papers 5 months ago

Class Attribute Inference Attacks: Inferring Sensitive Class Information by Diffusion-Based Attribute Manipulations

Paper • 2303.09289 • Published Mar 16, 2023 • 2

Distilling Adversarial Prompts from Safety Benchmarks: Report for the Adversarial Nibbler Challenge

Paper • 2309.11575 • Published Sep 20, 2023

MultiFusion: Fusing Pre-Trained Models for Multi-Lingual, Multi-Modal Image Generation

Paper • 2305.15296 • Published May 24, 2023 • 1

Mitigating Inappropriateness in Image Generation: Can there be Value in Reflecting the World's Ugliness?

Paper • 2305.18398 • Published May 28, 2023 • 2

Does CLIP Know My Face?

Paper • 2209.07341 • Published Sep 15, 2022 • 1

Exploiting Cultural Biases via Homoglyphs in Text-to-Image Synthesis

Paper • 2209.08891 • Published Sep 19, 2022 • 2

The Stable Artist: Steering Semantics in Diffusion Latent Space

Paper • 2212.06013 • Published Dec 12, 2022 • 1

LLavaGuard: VLM-based Safeguards for Vision Dataset Curation and Safety Assessment

Paper • 2406.05113 • Published Jun 7, 2024 • 3

AtMan: Understanding Transformer Predictions Through Memory Efficient Attention Manipulation

Paper • 2301.08110 • Published Jan 19, 2023 • 1