Feb 6 • 💬 General
Deep-Agent / R1-V: Reinforcing Super Generalization Ability in Vision Language Models with Less Than $3 We
"""
R1-V: Reinforcing Super Generalization Ability in Vision Language Models with Less Than $3
- We firstly reveal that Reinforcement Learning with Verifiable Rewards (RLVR) outperforms chain-of-thought supervised fine-tuning (CoT-SFT) in both effectiveness and out-of-distribution (OOD) robustness for vision language models.
- In our experiment, we incentivize VLMs to learn generalizable visual counting abilities, rather than overfitting to the training set.
- The 2B model outperforms the 72B model in OOD tests within just 100 training steps.
- The training was conducted on 8 A100 GPUs for 30 minutes, costing $2.62.
- Codes, models, datasets, more details and all open-source resources will be shared (within CNY holidays).
"""
1
1 comment
Deep-Agent / R1-V: Reinforcing Super Generalization Ability in Vision Language Models with Less Than $3 We
skool.com/data-alchemy
Your Community to Master the Fundamentals of Working with Data and AI — by Datalumina®
Leaderboard (30-day)
1
+203
2
+73
3
+32
4
+32
5
+29
Powered by