Unchun Yang's picture

Unchun Yang

ucyang

·

https://ucyang.com/

AI & ML interests

None yet

Recent Activity

liked a model about 1 hour ago

TinyLlama/TinyLlama_v1.1_math_code

upvoted a collection about 1 hour ago

AI PC: Text Generation

liked a dataset about 1 hour ago

argilla/distilabel-intel-orca-dpo-pairs

View all activity

Organizations

ucyang's activity

upvoted a collection about 1 hour ago

AI PC: Text Generation

Text generation LLMs that have been validated to run on the AI PC Intel® Core™ Ultra CPU and iGPU. • 186 items • Updated Aug 28 • 4

upvoted 2 collections 1 day ago

LLM2CLIP

LLM2CLIP makes SOTA pretrained CLIP modal more SOTA ever. • 10 items • Updated 15 days ago • 49

QVQ

QVQ: Qwen models for visual reasoning • 4 items • Updated 1 day ago • 19

upvoted a paper 4 days ago

Spectrum: Targeted Training on Signal to Noise Ratio

Paper • 2406.06623 • Published Jun 7 • 11

upvoted an article 5 days ago

Article

Selective fine-tuning of Language Models with Spectrum

By

•

Sep 3

• 30

upvoted 2 papers 5 days ago

Large Language Monkeys: Scaling Inference Compute with Repeated Sampling

Paper • 2407.21787 • Published Jul 31 • 12

Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling

Paper • 2412.05271 • Published 19 days ago • 121

upvoted 2 collections 5 days ago

InternVL2.5

Better than InternVL 2.0 • 18 items • Updated 5 days ago • 77

Stable Diffusion 3.5

6 items • Updated Oct 29 • 116

upvoted a paper 6 days ago

Qwen2.5 Technical Report

Paper • 2412.15115 • Published 6 days ago • 328

upvoted 5 collections 6 days ago

Llama 3.1 Collection

Meta's Llama 3.1 models including 8B, 70B, 405B. Includes 4-bit bnb and original versions. • 10 items • Updated 1 day ago • 1

Unsloth 4-bit Dynamic Quants

Unsloths Dynamic 4bit Quants selectively skips quantizing certain parameters; greatly improving accuracy while only using <10% more VRAM than BnB 4bit • 7 items • Updated 1 day ago • 20

Llama 3.3 (All Versions)

Meta's new Llama 3.3 (70B) model in all formats. Includes GGUF, 4-bit bnb and original versions. • 3 items • Updated 1 day ago • 27

SmolVLM

State-of-the-art compact VLMs for on-device applications: Base, Synthetic, and Instruct • 5 items • Updated 3 days ago • 30

📐 FineMath

FineMath datasets and ablation models • 13 items • Updated 3 days ago • 11

upvoted 2 papers 6 days ago

No More Adam: Learning Rate Scaling at Initialization is All You Need

Paper • 2412.11768 • Published 10 days ago • 41

Smarter, Better, Faster, Longer: A Modern Bidirectional Encoder for Fast, Memory Efficient, and Long Context Finetuning and Inference

Paper • 2412.13663 • Published 8 days ago • 103

upvoted a collection 6 days ago

ModernBERT

Bringing BERT into modernity via both architecture changes and scaling • 3 items • Updated 7 days ago • 93

upvoted a collection 9 days ago

Falcon3

Falcon3 family of Open Foundation Models is a set of pretrained and instruct LLMs ranging from 1B to 10B parameters. • 40 items • Updated 7 days ago • 71

upvoted a collection 11 days ago

SwiftKV Models

SwiftKV reduces prefill compute by up to 50% by combining model rewiring and knowledge-preserving self-distillation. • 3 items • Updated 21 days ago • 2