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 about 1 month ago
Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameters liked a model about 2 months ago
stepfun-ai/Step-3.5-Flash liked a dataset about 2 months ago
stepfun-ai/CF-Div2-StepfunOrganizations
models 8
RoadQAQ/ReLIFT-Qwen2.5-Math-7B-Zero
Question Answering • 8B • Updated • 5
RoadQAQ/Qwen2.5-7B-think
Text Generation • 8B • Updated
RoadQAQ/Qwen2.5-Math-1.5B-16k-think
Text Generation • 2B • Updated • 2.23k
RoadQAQ/ReLIFT-Qwen2.5-7B-Zero
Question Answering • 8B • Updated • 1 • 2
RoadQAQ/Qwen2.5-Math-7B-16k-think
Text Generation • 8B • Updated • 2
RoadQAQ/ReLIFT-Qwen2.5-Math-1.5B-Zero
Question Answering • 2B • Updated • 2
RoadQAQ/OpenR1-Distill-7B
Updated
RoadQAQ/video_llm_template
Updated