peak music, flawless performance ＼⁠(⁠^⁠o⁠^⁠)⁠／

🚨 Presenting today at #CVPR2026! Our highlight ⭐ paper Retrieve and Segment (RNS) asks a simple question: Can a few examples bridge the supervision gap in open-vocabulary segmentation? ✅ Turns out they can. 📍 Poster #578 ⏰ 16:45–18:45 📄 arxiv.org/abs/2602.23339

📍 @c1rcuslegend.bsky.social is presenting our work at #CVPR2026 Findings, find the poster with the most 🐈🐾 🗓 Friday, 07:00–08:30 📌 ExHall A; Poster #166 Stop by if you're curious whether MLLMs make good classifiers or to discuss our "Doomed to Reannotate" ImageNet project (preprint coming soon) 📝

To study this, we introduce ReGT, a new multilabel reannotation of 625 ImageNet classes that corrects many of these issues. When evaluated on the cleaned labels, multimodal LLMs improve by up to +10.8% accuracy, substantially narrowing the gap with supervised vision models. 📈