EviRover: Reinforcing Agentic Perception Beyond a Glance
[Widzenie komputerowe] Autorzy: Kaixuan Fan, Kaituo Feng, Tianshuo Peng i 4 więcej
Visual perception is conventionally formulated as a one-shot prediction from a single glance at the image, under the assumption that the image content and the model's parametric knowledge suffice to resolve the query. This assumption often fails in real-world scenarios that hinge on fine-grained visual details or require knowledge-intensive and up-to-date information. We term such cases \textit{perception under insufficient evidence} and formulate perception as an agentic process that can obtain information beyond a single glance. To address the absence of data for this setting, we design two dedicated data generation pipelines, yielding EviRover-SFT-5K and EviRover-RL-12K for training. We further construct EviLens, a human-verified benchmark comprising 688 instances across five perception categories. Building on these data, we present EviRover, to our knowledge the first perception age…
- Kategoria arxiv
- cs.CV
- Kategoria
- Widzenie komputerowe