ARM-Thinker: Reinforcing Multimodal Generative Reward Models with Agentic Tool Use and Visual Reasoning
By: Shengyuan Ding , Xinyu Fang , Ziyu Liu and more
Potential Business Impact:
Helps AI check its answers using tools.
Reward models are critical for aligning vision-language systems with human preferences, yet current approaches suffer from hallucination, weak visual grounding, and an inability to use tools for verification, limiting their reliability on complex multimodal reasoning tasks. We present ARM-Thinker, an A}gentic multimodal Reward Model that autonomously invokes external tools (e.g., image cropping, doc page retrieval) to ground judgments in verifiable evidence, replacing static, non-interactive reward scoring. This enables the model to verify fine-grained visual details, cross-reference multi-page evidence, and validate reasoning claims, which are capabilities absent in existing reward models. We train ARM-Thinker with multi-stage reinforcement learning, jointly optimizing tool-calling decisions and judgment accuracy. To evaluate agentic reward modeling, we introduce ARMBench-VL, comprising three benchmarks that assess fine-grained visual grounding (image-level tools), multi-page document understanding (retrieval tools), and instruction following (text-level verification). ARM-Thinker achieves +16.2% average improvement on reward modeling benchmarks, +9.6% on tool-use tasks, and outperforms baselines on multimodal math and logical reasoning benchmarks. Our results demonstrate that agentic capabilities significantly enhance both accuracy and interpretability of reward models.
Similar Papers
VR-Thinker: Boosting Video Reward Models through Thinking-with-Image Reasoning
CV and Pattern Recognition
Helps AI understand videos better by remembering details.
Multimodal Reinforcement Learning with Agentic Verifier for AI Agents
Artificial Intelligence
Teaches AI to think better and avoid mistakes.
One Model to Critique Them All: Rewarding Agentic Tool-Use via Efficient Reasoning
Artificial Intelligence
Helps AI use tools better and smarter.