Hugging Face
Models
Datasets
Spaces
Community
Docs
Enterprise
Pricing
Log In
Sign Up
36.4
TFLOPS
15
101
411
alkinun
AtAndDev
Follow
MongTe0712's profile picture
thomwolf's profile picture
not-lain's profile picture
51 followers
·
79 following
alkinun
alkinun
AI & ML interests
LLMs, Alignment, Merging, Unsloth, DPO, SFT, ORPO, SPIN..
Recent Activity
reacted
to
andito
's
post
with 🔥
about 24 hours ago
🧠👁️ Can AI visualize solutions? Humans often solve visual problems by sketching ideas in our minds. What if Vision-Language Models (VLMs) could do something similar, not by generating full images, but by using internal “mental sketches”? That’s the idea behind Mirage, a new framework that empowers VLMs to reason using latent visual tokens. Instead of just thinking in words, Mirage mixes in abstract visual representations that help the model solve complex tasks. These aren't photorealistic images. They're compact, internal representations optimized purely to support reasoning. 🔧 Mirage is trained in two phases: 1) Grounding: It learns to produce latent tokens anchored in real images. 2) Refinement: The model drops the images and learns to generate visual tokens on its own. 📈 And yes, it works! On challenging benchmarks like Visual Spatial Planning, Jigsaw puzzles, and Spatial Attention Tasks, Mirage clearly outperforms GPT-4o and other strong baselines. Smart sketches > empty words. By mimicking the way humans visualize solutions, Mirage gives AI a new kind of imagination, one that’s faster, more efficient, and more human-like. Kudos to the teams at UMass Amherst and MIT behind this exciting work. Check the paper: https://huggingface.co/papers/2506.17218
upvoted
a
paper
1 day ago
From Code to Correctness: Closing the Last Mile of Code Generation with Hierarchical Debugging
reacted
to
YerbaPage
's
post
with 🔥
1 day ago
Is 100% Pass Rate on HumanEval possible? Yes! ✅ Meet MGDebugger if you are tired of LLMs failing on complex bugs 🤔 Our MGDebugger, just hit 100% accuracy on HumanEval using the DeepSeek-R1 model. 🚀 ✨ Demo: https://huggingface.co/spaces/learnmlf/MGDebugger 📝 Paper: https://huggingface.co/papers/2410.01215 💻 Code: https://github.com/YerbaPage/MGDebugger HumanEval may be retired, we're ready for the next challenge In more complex scenarios! You may also take look at this repo for a collection of awesome repo-level coding tasks! 🖥️ https://github.com/YerbaPage/Awesome-Repo-Level-Code-Generation
View all activity
Organizations
Posts
9
view post
Post
2826
deepseek-ai/DeepSeek-R1-0528
This is the end
See translation
view post
Post
3087
Llama 4 is out...
View all Posts
spaces
3
Sort: Recently updated
Running
on
Zero
marco-qwq-7B
💻
Sleeping
AIDC AI Marco O1
💻
Generate responses for AI chat
Runtime error
Hallo
👋
Generate realistic talking heads from image+audio
models
9
Sort: Recently updated
AtAndDev/SelfCoder-v0.0
Text Generation
•
4B
•
Updated
May 19
•
5
AtAndDev/lora_model
Updated
Apr 5
AtAndDev/marco-qwq-7B
Text Generation
•
8B
•
Updated
Dec 8, 2024
•
12
AtAndDev/Ogno-Monarch-Neurotic-9B-Passthrough
Text Generation
•
9B
•
Updated
Mar 1, 2024
•
12
AtAndDev/Ogno-Monarch-Neurotic-7B-Dare-Ties
Text Generation
•
7B
•
Updated
Mar 1, 2024
•
13
AtAndDev/Marcoro14-7B-Slerp
Text Generation
•
7B
•
Updated
Mar 1, 2024
•
9
AtAndDev/CapybaraMarcoroni-7B
Text Generation
•
7B
•
Updated
Jan 7, 2024
•
848
AtAndDev/ShortKing-3b-v0.2
Text Generation
•
3B
•
Updated
Oct 2, 2023
•
85
•
2
AtAndDev/ShortKing-1.4b-v0.1
Text Generation
•
1B
•
Updated
Sep 29, 2023
•
3.17k
•
2
datasets
15
Sort: Recently updated
AtAndDev/SPRL-v0.1
Viewer
•
Updated
May 28
•
936
•
108
AtAndDev/SelfCoder-Test
Viewer
•
Updated
May 28
•
936
•
61
AtAndDev/ranky-dataset
Viewer
•
Updated
Mar 19
•
2.86k
•
33
AtAndDev/symbolm
Viewer
•
Updated
Jan 23
•
20k
•
46
AtAndDev/symlm
Viewer
•
Updated
Jan 16
•
10.1k
•
31
AtAndDev/chain-of-diffusion
Viewer
•
Updated
Jan 7
•
6.45k
•
35
AtAndDev/clip-bicycle-e-bike
Viewer
•
Updated
Jan 2
•
6k
•
61
AtAndDev/QwQ-LongCoT-59k-cleaned
Viewer
•
Updated
Dec 6, 2024
•
59.2k
•
46
•
1
AtAndDev/sedir-clean
Viewer
•
Updated
Dec 5, 2024
•
11.8k
•
38
AtAndDev/sedir-unclean
Viewer
•
Updated
Dec 5, 2024
•
19.9k
•
30
View 15 datasets