Through the Lens of Contrast: Self-Improving Visual Reasoning in VLMs Paper • 2603.02556 • Published Mar 3 • 4
VC-STaR Collection Official collection of models and datasets for the ICLR 2026 Oral paper "Through the Lens of Contrast: Self-Improving Visual Reasoning in VLMs" • 3 items • Updated 26 days ago • 2
Alibaba-Apsara/Superior-Reasoning-SFT-gpt-oss-120b-Logprob Viewer • Updated Jan 15 • 435k • 7.06k • 62