Checkpoints for the NeurIPS 2026 paper "Controllable Multi-label Video Safety Detection via Adaptive Tversky Policy Optimization".
-
BruceYg/ATPO-G-Qwen2.5-VL-7B
Reinforcement Learning • 8B • Updated • 26 -
BruceYg/ATPO-G-Qwen2.5-VL-7B-r5
Reinforcement Learning • 8B • Updated • 23 -
BruceYg/ATPO-G-Qwen2.5-VL-7B-r0.2
Reinforcement Learning • 8B • Updated • 26 • 1 -
BruceYg/ATPO-C-Qwen2.5-VL-7B-C4r5-C6r0.2
Reinforcement Learning • 8B • Updated • 14