Chenyang Song

Raincleared's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

upvoted a paper 19 days ago

Uni-MoE-2.0-Omni: Scaling Language-Centric Omnimodal Large Model with Advanced MoE, Training and Data

Paper • 2511.12609 • Published 21 days ago • 102

upvoted 2 papers 4 months ago

Speed Always Wins: A Survey on Efficient Architectures for Large Language Models

Paper • 2508.09834 • Published Aug 13 • 53

μ-Parametrization for Mixture of Experts

Paper • 2508.09752 • Published Aug 13 • 10

upvoted a paper 5 months ago

BlockFFN: Towards End-Side Acceleration-Friendly Mixture-of-Experts with Chunk-Level Activation Sparsity

Paper • 2507.08771 • Published Jul 11 • 9

upvoted a collection 6 months ago

MiniCPM4

Collection

MiniCPM4: Ultra-Efficient LLMs on End Devices • 29 items • Updated Sep 8 • 79

upvoted a paper 7 months ago

The Sparse Frontier: Sparse Attention Trade-offs in Transformer LLMs

Paper • 2504.17768 • Published Apr 24 • 13

upvoted a paper 9 months ago

SEAP: Training-free Sparse Expert Activation Pruning Unlock the Brainpower of Large Language Models

Paper • 2503.07605 • Published Mar 10 • 68

upvoted 3 papers about 1 year ago

Densing Law of LLMs

Paper • 2412.04315 • Published Dec 5, 2024 • 19

Sparsing Law: Towards Large Language Models with Greater Activation Sparsity

Paper • 2411.02335 • Published Nov 4, 2024 • 11

Configurable Foundation Models: Building LLMs from a Modular Perspective

Paper • 2409.02877 • Published Sep 4, 2024 • 31

upvoted 3 papers over 1 year ago

Beyond the Turn-Based Game: Enabling Real-Time Conversations with Duplex Models

Paper • 2406.15718 • Published Jun 22, 2024 • 14

Turbo Sparse: Achieving LLM SOTA Performance with Minimal Activated Parameters

Paper • 2406.05955 • Published Jun 10, 2024 • 27

PowerInfer-2: Fast Large Language Model Inference on a Smartphone

Paper • 2406.06282 • Published Jun 10, 2024 • 38

upvoted 2 collections over 1 year ago

MiniCPM

Collection

The MiniCPM family of LLMs and VLLMs. • 33 items • Updated Aug 7 • 73

Meta Llama 3

Collection

This collection hosts the transformers and original repos of the Meta Llama 3 and Llama Guard 2 releases • 5 items • Updated Dec 6, 2024 • 872

upvoted 5 papers almost 2 years ago

In deep reinforcement learning, a pruned network is a good network

Paper • 2402.12479 • Published Feb 19, 2024 • 19

LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens

Paper • 2402.13753 • Published Feb 21, 2024 • 116

SliceGPT: Compress Large Language Models by Deleting Rows and Columns

Paper • 2401.15024 • Published Jan 26, 2024 • 74

DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models

Paper • 2401.06066 • Published Jan 11, 2024 • 58

Mixtral of Experts

Paper • 2401.04088 • Published Jan 8, 2024 • 160