Haoyuan WU

hywu's activity

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

upvoted 2 papers 1 day ago

Nested Learning: The Illusion of Deep Learning Architectures

Paper • 2512.24695 • Published 7 days ago • 27

Diversity or Precision? A Deep Dive into Next Token Prediction

Paper • 2512.22955 • Published 9 days ago • 5

upvoted a paper 2 months ago

Knocking-Heads Attention

Paper • 2510.23052 • Published Oct 27, 2025 • 29

upvoted 2 papers 3 months ago

One-Token Rollout: Guiding Supervised Fine-Tuning of LLMs with Policy Gradient

Paper • 2509.26313 • Published Sep 30, 2025 • 4

Reinforcement Learning on Pre-Training Data

Paper • 2509.19249 • Published Sep 23, 2025 • 67

upvoted a collection 5 months ago

GroveMoE

Collection

3 items • Updated Aug 22, 2025 • 1

upvoted a paper 5 months ago

Parameter-Efficient Sparsity Crafting from Dense to Mixture-of-Experts for Instruction Tuning on General Tasks

Paper • 2401.02731 • Published Jan 5, 2024 • 3

upvoted a collection 5 months ago

GroveMoE

Collection

GroveMoE is an open-source family of large language models developed by the AGI Center, Ant Research Institute. • 4 items • Updated 13 days ago • 7

upvoted a paper 5 months ago

Grove MoE: Towards Efficient and Superior MoE LLMs with Adjugate Experts

Paper • 2508.07785 • Published Aug 11, 2025 • 28

upvoted an article 11 months ago

view article

Article

Putting RL back in RLHF

• 
• 

Jun 12, 2024

•

109

upvoted a collection 11 months ago

Camelidae

Collection

5 items • Updated Aug 22, 2025 • 2

upvoted a collection 12 months ago

Cosmos

Collection

⚠️ This collection is archived. 👉 https://huggingface.co/collections/nvidia/nvidia-cosmos-2 • 31 items • Updated 1 day ago • 299

upvoted a collection about 1 year ago

Tulu 3 Datasets

Collection

All datasets released with Tulu 3 -- state of the art open post-training recipes. • 33 items • Updated 14 days ago • 96