Minghan Wang

wangminghan's activity

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

upvoted a collection about 6 hours ago

Scheduling Recursive Reasoning in Looped Transformers

Collection

TRM checkpoints trained with progress–stability co-design for TAPS adaptive recurrent inference. Includes Sudoku and Maze models, config • 3 items • Updated 1 day ago • 5

upvoted a paper about 17 hours ago

Thinking Outside the Box: Can Language Models Rely on External Guidance Selectively?

Paper • 2609.39578 • Published 2 days ago • 45

upvoted a collection 1 day ago

Thinking Outside the Box

Collection

This repository provides Thinking Outside the Box-trained models for mathematical reasoning and WebShop tasks. It includes the SFT and RL checkpoints. • 6 items • Updated about 17 hours ago • 2