Xiangyu

xixy's activity

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

commented a paper 11 days ago

Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B

Paper • 2511.06221 • Published 15 days ago • 114 •

11

commented a paper 12 days ago

DRIVE: Data Curation Best Practices for Reinforcement Learning with Verifiable Reward in Competitive Code Generation

Paper • 2511.06307 • Published 14 days ago • 50 •

5

upvoted a paper 12 days ago

DRIVE: Data Curation Best Practices for Reinforcement Learning with Verifiable Reward in Competitive Code Generation

Paper • 2511.06307 • Published 14 days ago • 50

upvoted an article 13 days ago

view article

Article

Budget Alignment: Making Models Reason in the User’s Language

• 

19 days ago

•

9

upvoted a paper 23 days ago

AMO-Bench: Large Language Models Still Struggle in High School Math Competitions

Paper • 2510.26768 • Published 24 days ago • 33

commented 3 papers about 1 month ago

Learning from the Best, Differently: A Diversity-Driven Rethinking on Data Selection

Paper • 2510.18909 • Published Oct 21 • 4 •

3

First Try Matters: Revisiting the Role of Reflection in Reasoning Models

Paper • 2510.08308 • Published Oct 9 • 24 •

4

First Try Matters: Revisiting the Role of Reflection in Reasoning Models

Paper • 2510.08308 • Published Oct 9 • 24 •

4

upvoted 2 papers about 1 month ago

VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applications

Paper • 2509.26490 • Published Sep 30 • 19

R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?

Paper • 2510.08189 • Published Oct 9 • 26

New activity in XenArcAI/MathX-5M 4 months ago

What is the model used to produce responses?

5

#3 opened 4 months ago by

xixy

upvoted a collection 4 months ago

OpenReasoning-Nemotron

Collection

Collection of models for OpenReasoning-Nemotron which are trained on 5M reasoning traces for Math, Code and Science. • 6 items • Updated 2 days ago • 44

upvoted a paper 5 months ago

AceReason-Nemotron 1.1: Advancing Math and Code Reasoning through SFT and RL Synergy

Paper • 2506.13284 • Published Jun 16 • 26

New activity in a-m-team/AM-DeepSeek-R1-0528-Distilled 5 months ago

什么叫中国速度！

6

#1 opened 6 months ago by

reign12

commented a paper 6 months ago

Reinforcement Pre-Training

Paper • 2506.08007 • Published Jun 9 • 262 •

49

New activity in a-m-team/AM-DeepSeek-R1-0528-Distilled 6 months ago

请问有code相关的评测结果吗？

#2 opened 6 months ago by

xixy

commented a paper 6 months ago

Why Distillation can Outperform Zero-RL: The Role of Flexible Reasoning

Paper • 2505.21067 • Published May 27 • 3 •

1

authored a paper 6 months ago

Rethinking the Sampling Criteria in Reinforcement Learning for LLM Reasoning: A Competence-Difficulty Alignment Perspective

Paper • 2505.17652 • Published May 23 • 6

upvoted 2 papers 6 months ago

Rethinking the Sampling Criteria in Reinforcement Learning for LLM Reasoning: A Competence-Difficulty Alignment Perspective

Paper • 2505.17652 • Published May 23 • 6

Not All Correct Answers Are Equal: Why Your Distillation Source Matters

Paper • 2505.14464 • Published May 20 • 9