Dmitry Ryumin's picture

Dmitry Ryumin

DmitryRyumin

·

https://dmitryryumin.github.io

DmitryRyumin

AI & ML interests

Machine Learning and Applications, Multi-Modal Understanding

Recent Activity

liked a model 6 days ago

XiaomiMiMo/MiMo-7B-RL

reacted to merterbak's post with 🔥 7 days ago

Qwen 3 models released🔥 It offers 2 MoE and 6 dense models with following parameter sizes: 0.6B, 1.7B, 4B, 8B, 14B, 30B(MoE), 32B, and 235B(MoE). Models: https://huggingface.co/collections/Qwen/qwen3-67dd247413f0e2e4f653967f Blog: https://qwenlm.github.io/blog/qwen3/ Demo: https://huggingface.co/spaces/Qwen/Qwen3-Demo GitHub: https://github.com/QwenLM/Qwen3 ✅ Pre-trained 119 languages(36 trillion tokens) and dialects with strong translation and instruction following abilities. (Qwen2.5 was pre-trained on 18 trillion tokens.) ✅Qwen3 dense models match the performance of larger Qwen2.5 models. For example, Qwen3-1.7B/4B/8B/14B/32B perform like Qwen2.5-3B/7B/14B/32B/72B. ✅ Three stage done while pretraining: • Stage 1: General language learning and knowledge building. • Stage 2: Reasoning boost with STEM, coding, and logic skills. • Stage 3: Long context training ✅ It supports MCP in the model ✅ Strong agent skills ✅ Supports seamless between thinking mode (for hard tasks like math and coding) and non-thinking mode (for fast chatting) inside chat template. ✅ Better human alignment for creative writing, roleplay, multi-turn conversations, and following detailed instructions.

upvoted a collection 7 days ago

View all activity

Organizations

DmitryRyumin's activity

upvoted 2 collections 7 days ago

Qwen3

27 items • Updated 4 days ago • 515

MAI-DS-R1

MAI-DS-R1 is a DeepSeek-R1 reasoning model that has been post-trained by the Microsoft AI team. • 2 items • Updated 5 days ago • 11

upvoted a collection 8 days ago

Kimi-Audio-7B

Kimi audio 7B models • 3 items • Updated 8 days ago • 8

upvoted a collection 9 days ago

Gemma 3 Release

24 items • Updated 18 days ago • 355

upvoted an article 11 days ago

Article

Welcome the Falcon 3 Family of Open Models!

Dec 17, 2024

• 127

upvoted a paper 15 days ago

HunyuanPortrait: Implicit Condition Control for Enhanced Portrait Animation

Paper • 2503.18860 • Published Mar 24 • 3

upvoted a paper 17 days ago

Reconstructing Humans with a Biomechanically Accurate Skeleton

Paper • 2503.21751 • Published Mar 27 • 9

upvoted a paper 19 days ago

EmoAgent: Assessing and Safeguarding Human-AI Interaction for Mental Health Safety

Paper • 2504.09689 • Published 22 days ago • 7

upvoted a collection 25 days ago

heartsync-mbti

16 items • Updated 25 days ago • 59

upvoted 2 papers 25 days ago

FantasyTalking: Realistic Talking Portrait Generation via Coherent Motion Synthesis

Paper • 2504.04842 • Published 29 days ago • 36

DDT: Decoupled Diffusion Transformer

Paper • 2504.05741 • Published 28 days ago • 73

upvoted a paper 28 days ago

TransMamba: Flexibly Switching between Transformer and Mamba

Paper • 2503.24067 • Published Mar 31 • 20

upvoted an article 29 days ago

Article

Welcome Llama 4 Maverick & Scout on Hugging Face!

Apr 5

• 142

upvoted a paper about 1 month ago

Multi-Token Attention

Paper • 2504.00927 • Published Apr 1 • 49

upvoted a collection about 1 month ago

MambaVision

MambaVision: A Hybrid Mamba-Transformer Vision Backbone. Includes both 1K and 21K pretrained models. • 13 items • Updated about 8 hours ago • 31

upvoted 2 papers about 1 month ago

TaoAvatar: Real-Time Lifelike Full-Body Talking Avatars for Augmented Reality via 3D Gaussian Splatting

Paper • 2503.17032 • Published Mar 21 • 26

MAPS: A Multi-Agent Framework Based on Big Seven Personality and Socratic Guidance for Multimodal Scientific Problem Solving

Paper • 2503.16905 • Published Mar 21 • 54

upvoted 2 papers about 2 months ago

Transformers without Normalization

Paper • 2503.10622 • Published Mar 13 • 162

Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders

Paper • 2503.03601 • Published Mar 5 • 232

upvoted a collection 2 months ago

C2SER

3 items • Updated Feb 25 • 2