Ctrl+K
- Qwen2.5-VL-7B-Instruct
- Qwen2.5-VL-7B-Proposer-Step1
- Qwen2.5-VL-7B-Proposer-Step2-GDPO-Stability
- Qwen2.5-VL-7B-Proposer-Step2-GDPO
- Qwen2.5-VL-7B-Proposer-Step2-GDPOFULL
- Qwen2.5-VL-7B-Proposer-Step2-wo-feedback
- Qwen2.5-VL-7B-Proposer-Step2
- Qwen2.5-VL-7B-Proposer-Step3-FULL-Softmax-thd0_5
- Qwen2.5-VL-7B-Proposer-Step3-FULL-Softmax
- Qwen2.5-VL-7B-Proposer-Step3-GDPO-Stability
- Qwen2.5-VL-7B-Proposer-Step3-GDPO
- Qwen2.5-VL-7B-Proposer-Step3-GDPOFULL
- Qwen2.5-VL-7B-Proposer-Step3-wo-feedback
- Qwen2.5-VL-7B-Proposer-Step3
- Step2-RZero-Style
- Step3-FULL-Softmax-thd0_5-FT-Charades
- Step3-FULL-Softmax-thd0_5
- 1.52 kB