Hugging Face's logo

Social Post Explorers
community
Activity Feed

AI & ML interests

None defined yet.

Recent Activity

DmitryRyumin  authored a paper about 1 month ago
Team RAS in 9th ABAW Competition: Multimodal Compound Expression Recognition Approach
stefan-it  authored a paper about 1 month ago
SindBERT, the Sailor: Charting the Seas of Turkish NLP
stefan-it  authored a paper about 2 months ago
The German Commons - 154 Billion Tokens of Openly Licensed Text for German Language Models
View all activity

leonardlin 
posted an update about 5 hours ago
view post
Post
76
We just released our latest Shisa V2.1 Japanese multi-lingual models: https://huggingface.co/collections/shisa-ai/shisa-v21

Besides updates to our 14B, and 70B, we have a new LFM2-based 1.2B, Llama 3.2-based 3B, and Qwen 3-based 8B, all with class-leading Japanese language capabilities.

Per usual, lots of details in the Model Cards for those interested.
fernandofernandes 
authored 4 papers 8 days ago
juhoinkinen 
posted an update 13 days ago
view post
Post
227
**AI4LAM’s annual conference, AI Everywhere, All at Once
December 3 – 5, 2025, British Library, London**

See the conference programme: 👉 https://www.conftool.org/fantastic-futures-2025/sessions.php

Some program items related to NatLibFi/Annif:
• Workshop:
• Evaluating Automated Subject Indexing Methods, Maximilian Kähler
• Presentations:
• Autocat Cataloguing Assistant
• The usage of hardware resources for automatic subject cataloguing at the German National Library – an analysis and outlook for future challenges, Christoph Poley
• Posters:
• AI-Powered Subject Indexing in the Archives – Piloting Finto AI at the Finnish Literature Society, Milla Eräsaari and Teemu Hirvonen
• From Annotation to Insight: Human-in-the-Loop Machine Learning for Historical Archives in HAICu WP2, C.A. Romein and others
Bils 
posted an update 26 days ago
xianbao 
authored a paper about 1 month ago
nouamanetazi 
posted an update about 1 month ago
view post
Post
3972
After training 𝐒𝐦𝐨𝐥𝐋𝐌𝟑 on 𝟑𝟖𝟒 𝐇𝟏𝟎𝟎𝐬 for nearly a month, I've come to realize something most people overlook: 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐢𝐬 𝐭𝐡𝐞 𝐦𝐚𝐤𝐞-𝐨𝐫-𝐛𝐫𝐞𝐚𝐤 𝐟𝐚𝐜𝐭𝐨𝐫 𝐢𝐧 𝐋𝐋𝐌 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠. 🔥

Everyone talks about model architecture and data quality. And yes, those matter immensely. But here's what nobody tells you: when your training run fails at 2 AM because of mysterious 𝐍𝐂𝐂𝐋 𝐞𝐫𝐫𝐨𝐫𝐬, or when your expensive GPU cluster is running at 𝟔𝟎% 𝐞𝐟𝐟𝐢𝐜𝐢𝐞𝐧𝐜𝐲, the problem isn't your model. It's most probably a 𝐦𝐢𝐬𝐮𝐬𝐞 𝐨𝐟 𝐭𝐡𝐞 𝐡𝐚𝐫𝐝𝐰𝐚𝐫𝐞. 🛠️

Questions that seemed simple but had no clear answers: Why is 𝐌𝐨𝐄 𝐭𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐬𝐥𝐨𝐰𝐞𝐫 𝐭𝐡𝐚𝐧 𝐝𝐞𝐧𝐬𝐞 𝐦𝐨𝐝𝐞𝐥𝐬? Which 𝐍𝐂𝐂𝐋 𝐟𝐥𝐚𝐠𝐬 should we actually set? How often should we checkpoint without killing throughput?

That's why we built 𝐓𝐡𝐞 𝐒𝐦𝐨𝐥 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐲𝐛𝐨𝐨𝐤 📖: a complete guide covering everything from model architecture and data curation to the SmolLM3 training marathon, post-training techniques, and crucially, the 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 𝐥𝐚𝐲𝐞𝐫 that most teams get wrong.

We validated real vs theoretical bandwidth across the entire stack: 𝐇𝐁𝐌𝟑 𝐡𝐢𝐭𝐭𝐢𝐧𝐠 𝟑 𝐓𝐁/𝐬, 𝐍𝐕𝐋𝐢𝐧𝐤 𝟒.𝟎 𝐫𝐞𝐚𝐜𝐡𝐢𝐧𝐠 𝟕𝟖𝟔 𝐆𝐁/𝐬, 𝐏𝐂𝐈𝐞 𝐆𝐞𝐧𝟒 𝐚𝐭 𝟏𝟒.𝟐 𝐆𝐁/𝐬. Then we ran collective operations across 𝟏𝟐𝟖 𝐆𝐏𝐔𝐬 (16 nodes, 8xH100s each) and measured how performance degrades at scale: all-reduce drops from 𝟒𝟖𝟎 𝐆𝐁/𝐬 on a single node to 𝟑𝟐𝟎-𝟑𝟓𝟎 𝐆𝐁/𝐬 across 16 nodes.

If you've ever wondered why your training runs are slower than they should be, or you're planning to scale up and want to avoid expensive mistakes, this guide might save you weeks of debugging.

𝐓𝐡𝐞 𝐒𝐦𝐨𝐥 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐲𝐛𝐨𝐨𝐤: https://lnkd.in/e5MKXUHS

Shared with ❤️ by the HuggingFace team
sam-paech 
authored 3 papers about 2 months ago
merve 
posted an update about 2 months ago
view post
Post
6667
deepseek-ai/DeepSeek-OCR is out! 🔥 my take ⤵️
> pretty insane it can parse and re-render charts in HTML
> it uses CLIP and SAM features concatenated, so better grounding
> very efficient per vision tokens/performance ratio
> covers 100 languages
·
lamhieu 
posted an update about 2 months ago
view post
Post
2735
🚀 Introducing the xLLMs Dataset Collection

The xLLMs project is a growing suite of multilingual and multimodal dialogue datasets designed to train and evaluate advanced conversational LLMs. Each dataset focuses on a specific capability — from long-context reasoning and factual grounding to STEM explanations, math Q&A, and polite multilingual interaction.

🌍 Explore the full collection on Hugging Face:
👉 lamhieu/xllms-66cdfe34307bb2edc8c6df7d

💬 Highlight: xLLMs – Dialogue Pubs
A large-scale multilingual dataset built from document-guided synthetic dialogues (Wikipedia, WikiHow, and technical sources). It’s ideal for training models on long-context reasoning, multi-turn coherence, and tool-augmented dialogue across 9 languages.
👉 lamhieu/xllms_dialogue_pubs

🧠 Designed for:
- Long-context and reasoning models
- Multilingual assistants
- Tool-calling and structured response learning

All datasets are open for research and development use — free, transparent, and carefully curated to improve dialogue model quality.
·
s3nh 
posted an update about 2 months ago
view post
Post
538
Eduhelp with more empathy, based on model finetuned on
psychotheraputic preferences just landed on


Beck-8B as a base model, 13000 steps on educational dataset.
Time to go further and build more 🥰
s3nh/EduHelp_Beck_8B
Thanks to @basilic_ai for computations <3
s3nh 
posted an update about 2 months ago
view post
Post
4107
Just tried to create an educational assistant for younger people who can struggle with visualsation of 'what is this sorcery all about'.
Its first step of my spare time projects, sft on Qwen3-8B,

EduHelper is a child-friendly tutoring assistant fine-tuned from the Qwen3-8B base model using parameter-efficient fine-tuning (PEFT) with LoRA on the ajibawa-2023/Education-Young-Children dataset.

s3nh/EduHelp-8B

Glad to share my work, have a wonderful day!
  • 2 replies
·
osanseviero 
authored a paper 3 months ago
merve 
posted an update 3 months ago
view post
Post
6714
large AI labs open-sourced a ton of models last week 🔥
here's few picks, find even more here merve/sep-16-releases-68d13ea4c547f02f95842f05 🤝
> IBM released a new Docling model with 258M params based on Granite (A2.0) 📝 ibm-granite/granite-docling-258M
> Xiaomi released 7B audio LM with base and instruct variants (MIT) XiaomiMiMo/mimo-audio-68cc7202692c27dae881cce0
> DecartAI released Lucy Edit, open Nano Banana 🍌 (NC) decart-ai/Lucy-Edit-Dev
> OpenGVLab released a family of agentic computer use models (3B/7B/32B) with the dataset 💻 OpenGVLab/scalecua-68c912cf56f7ff4c8e034003
> Meituan Longcat released thinking version of LongCat-Flash 💭 meituan-longcat/LongCat-Flash-Thinking
  • 2 replies
·
Tonic 
posted an update 3 months ago
merve 
posted an update 3 months ago
view post
Post
3337
IBM just released small swiss army knife for the document models: granite-docling-258M on Hugging Face 🔥

> not only a document converter but also can do document question answering, understand multiple languages 🤯
> best part: released with Apache 2.0 license 👏 use it with your commercial projects!
> it supports transformers, vLLM and MLX from the get-go! 🤗
> built on SigLIP2 & granite-165M

model: ibm-granite/granite-docling-258M
demo: ibm-granite/granite-docling-258m-demo 💗