ReLIFT, a training method that interleaves RL with online FT, achieving superior performance and efficiency compared to using RL or SFT alone.
RoadMa
RoadQAQ
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
1 day ago
Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters
liked
a model
4 days ago
stepfun-ai/Step-3.5-Flash
liked
a dataset
4 days ago
stepfun-ai/CF-Div2-Stepfun
Organizations
models
8
RoadQAQ/ReLIFT-Qwen2.5-Math-7B-Zero
Question Answering
•
8B
•
Updated
•
83
RoadQAQ/Qwen2.5-7B-think
Text Generation
•
8B
•
Updated
•
2
RoadQAQ/Qwen2.5-Math-1.5B-16k-think
Text Generation
•
2B
•
Updated
•
2.22k
RoadQAQ/ReLIFT-Qwen2.5-7B-Zero
Question Answering
•
8B
•
Updated
•
1
•
2
RoadQAQ/Qwen2.5-Math-7B-16k-think
Text Generation
•
8B
•
Updated
•
17
RoadQAQ/ReLIFT-Qwen2.5-Math-1.5B-Zero
Question Answering
•
2B
•
Updated
•
75
RoadQAQ/OpenR1-Distill-7B
Updated
RoadQAQ/video_llm_template
Updated