SpatialThinker: Reinforcing 3D Reasoning in Multimodal LLMs via Spatial Rewards

Hunar Batra, Haoqin Tu, Hardy Chen, Yuanze Lin, Cihang Xie, Ronald Clark

Published: 22 Sept 2025, Last Modified: 03 Jan 2026WiML @ NeurIPS 2025EveryoneRevisionsBibTeXCC BY 4.0
Keywords: Multimodal Large Language Models, RL, Spatial Reasoning, Reward Shaping, Dense Rewards
Submission Number: 372