Verifiable Visual Rewards Transfer from Synthetic Scenes to Natural Prompts
[Sztuczna inteligencja] Autorzy: Shuyue Stella Li, Xiaochuang Han, Yulia Tsvetkov i 1 więcej
Precise instruction following in image generation, such as satisfying object counts and spatial relations, remains an open challenge at least in part because it is learned using unreliable reward models such as object detectors and vision-language models. We introduce Verifiable Visual Rewards (VVR), the first framework for programmatically verifiable image rewards, and show that training on it generalizes to natural prompts. Each VVR task is a scene of geometric objects and relations among them, from which we derive both the prompt and a deterministic verifier, so tasks can be generated in any number and at any chosen complexity. We release VVRBench, with 10,000 tasks over 32 constraint types, and VVRBench-Challenge, with 720 more complex tasks; the strongest model we evaluate---GPT-Image-2.5---solves 21.4% of VVRBench-Challenge. Using VVR scores as rewards for reinforcement learning (…
- Kategoria arxiv
- cs.AI
- Kategoria
- Sztuczna inteligencja