Not Lain's picture

Building on HF

Not Lain

not-lain

chonkie-ai

·

https://not-lain.github.io

AI & ML interests

custom AI models with HF integration, HuggingFace fellow 🤗

Recent Activity

upvoted a paper 5 days ago

ShadowPEFT: Shadow Network for Parameter-Efficient Fine-Tuning

reacted to SeanLee97's post with 🚀 5 days ago

Our lab recently released a paper where we introduce ShadowPEFT, a new Parameter-Efficient Fine-Tuning (PEFT) paradigm tailored for edge computing scenarios. Unlike traditional approaches such as LoRA and its variants, which inject trainable parameters directly into the weights of Transformer, requiring tight coupling with the backbone. ShadowPEFT instead enhances the frozen large base model by adding a lightweight, centralized, pretrainable, and detachable Shadow network. This shadow network operates in parallel with the base model, delivering learned corrections to each decoder layer. Because the shadow module is architecturally decoupled from the backbone, it can be independently trained, stored, and deployed, benefiting edge computing scenarios and edge-cloud collaboration computing. - HF Paper: https://huggingface.co/papers/2604.19254 - GitHub: https://github.com/ShadowLLM/shadow-peft - HF Collection: https://huggingface.co/collections/shadow-llm/shadow-peft-models

reacted to SeanLee97's post with 🔥 5 days ago

Our lab recently released a paper where we introduce ShadowPEFT, a new Parameter-Efficient Fine-Tuning (PEFT) paradigm tailored for edge computing scenarios. Unlike traditional approaches such as LoRA and its variants, which inject trainable parameters directly into the weights of Transformer, requiring tight coupling with the backbone. ShadowPEFT instead enhances the frozen large base model by adding a lightweight, centralized, pretrainable, and detachable Shadow network. This shadow network operates in parallel with the base model, delivering learned corrections to each decoder layer. Because the shadow module is architecturally decoupled from the backbone, it can be independently trained, stored, and deployed, benefiting edge computing scenarios and edge-cloud collaboration computing. - HF Paper: https://huggingface.co/papers/2604.19254 - GitHub: https://github.com/ShadowLLM/shadow-peft - HF Collection: https://huggingface.co/collections/shadow-llm/shadow-peft-models

View all activity

Organizations

upvoted a paper 5 days ago

ShadowPEFT: Shadow Network for Parameter-Efficient Fine-Tuning

Paper • 2604.19254 • Published 6 days ago • 28

upvoted a changelog 7 days ago

Hugging Face Changelog

Introducing Kernels

12 days ago

• 150

upvoted an article 12 days ago

Article

Any Custom Frontend with Gradio's Backend

26 days ago

•

25

upvoted an article 13 days ago

Article

Community Evals: Because we're done trusting black-box leaderboards over the community

+5

Feb 4

•

89

upvoted an article 24 days ago

Article

Welcome Gemma 4: Frontier multimodal intelligence on device

+5

25 days ago

•

879

upvoted an article 28 days ago

Article

How I contributed a new model to the Transformers library using Codex

28 days ago

•

49

upvoted 2 changelogs about 1 month ago

Hugging Face Changelog

Protected Spaces with Public URLs

Mar 20

• 122

Hugging Face Changelog

Introducing hf-mount

Mar 24

• 223

upvoted 2 papers about 1 month ago

ObjectClear: Complete Object Removal via Object-Effect Attention

Paper • 2505.22636 • Published May 28, 2025 • 4

MatAnyone 2: Scaling Video Matting via a Learned Quality Evaluator

Paper • 2512.11782 • Published Dec 12, 2025 • 3

upvoted a changelog about 2 months ago

Hugging Face Changelog

Public Storage Add-ons

Feb 26

• 168

upvoted 2 articles 2 months ago

Article

GGML and llama.cpp join HF to ensure the long-term progress of Local AI

+4

Feb 20

•

504

Article

How to Use Multiple GPUs in Hugging Face Transformers: Device Map vs Tensor Parallelism

Feb 12

•

20

upvoted an article 3 months ago

Article

Introducing Daggr: Chain apps programmatically, inspect visually

+3

Jan 29

•

106

upvoted a collection 3 months ago

Trinity-Large

8 items • Updated 28 days ago • 42

upvoted a changelog 3 months ago

Hugging Face Changelog

MLX Hardware Compatibility

Jan 22

• 48

upvoted an article 3 months ago

Article

Small Yet Mighty: Improve Accuracy In Multimodal Search and Visual Document Retrieval with Llama Nemotron RAG Models

Jan 6

•

28

upvoted a paper 4 months ago

LTX-2: Efficient Joint Audio-Visual Foundation Model

Paper • 2601.03233 • Published Jan 6 • 177

upvoted an article 4 months ago

Article

Scaling Real-Time Voice Agents with Cache-Aware Streaming ASR

Jan 5

•

86

upvoted a changelog 4 months ago

Hugging Face Changelog

HuggingChat for Papers

Jan 7

• 103