The collection for the Paper "The Surprising Effectiveness of Negative Reinforcement in LLM Reasoning"
Xinyu Zhu
TianHongZXY
AI & ML interests
Large Language Models; Reasoning; Reinforcement Learning
Recent Activity
upvoted
a
paper
about 10 hours ago
ProFit: Leveraging High-Value Signals in SFT via Probability-Guided Token Selection
upvoted
a
paper
about 1 month ago
Wan-Move: Motion-controllable Video Generation via Latent Trajectory Guidance
updated
a model
about 1 month ago
TianHongZXY/Qwen3-4B-NSR
Organizations
models
13
TianHongZXY/Qwen3-4B-NSR
4B
•
Updated
•
2
TianHongZXY/Qwen3-4B-Thinking-2507-SFT-10-epochs-synthesized-clear-problems-global_step_280
0.5B
•
Updated
TianHongZXY/Qwen2.5-Math-7B-GRPO
8B
•
Updated
TianHongZXY/OpenR1-Math-46k-8192-Qwen2.5-Math-7B-RoPE-40K-GRPO-use_guide-clip_ratio_upper_0.28
Updated
TianHongZXY/OpenR1-Math-46k-8192-Qwen2.5-7B-Instruct-GRPO-clip_0.28
Updated
TianHongZXY/OpenR1-Math-46k-8192-Qwen2.5-7B-Instruct-GRPO-gpt-4o-summary_wo_think-clip_0.28
Updated
TianHongZXY/Qwen2.5-Math-7B-W-REINFORCE
8B
•
Updated
•
4
•
1
TianHongZXY/Qwen3-4B-GRPO
4B
•
Updated
•
2
TianHongZXY/Qwen3-4B-PPO
4B
•
Updated
TianHongZXY/Qwen3-4B-PSR
4B
•
Updated
•
14
•
1
datasets
58
TianHongZXY/similar_problems_with_three_in_context_problems
Viewer
•
Updated
•
2.16k
•
4.88k
TianHongZXY/Top_5_similar_question-NVIDIA-OpenScienceReasoning-2
Viewer
•
Updated
•
2.16k
•
3.71k
TianHongZXY/aime-1983-2025
Viewer
•
Updated
•
963
•
122
TianHongZXY/AIME2025
Viewer
•
Updated
•
30
•
442
•
1
TianHongZXY/AIME2024
Viewer
•
Updated
•
30
•
116
TianHongZXY/amc23
Viewer
•
Updated
•
40
•
301
TianHongZXY/MATH-test-Tulu-3-8B-SFT-beam_search-completions-temp_0.8-range_1500_to_1600
Viewer
•
Updated
•
100
•
5
TianHongZXY/MATH-test-Tulu-3-8B-SFT-beam_search-completions-temp_0.8-range_4900_to_5000
Viewer
•
Updated
•
100
•
1
TianHongZXY/MATH-test-Tulu-3-8B-SFT-beam_search-completions-temp_0.8-range_4800_to_4900
Viewer
•
Updated
•
100
TianHongZXY/MATH-test-Tulu-3-8B-SFT-beam_search-completions-temp_0.8-range_4700_to_4800
Viewer
•
Updated
•
100
•
2