AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

MindPower: Enabling Theory-of-Mind Reasoning in VLM-based Embodied Agents

Ruoxuan Zhang, Qiyun Zheng, Zhiyu Zhou, Ziqi Liao, Siyu Wu, Jian-Yu Jiang-Lin, Bin Wen, Hongxia Xie, Jianlong Fu, Wen-Huang Cheng

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Embodied multimodal LLM agents · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add SFT on the MindPower Reasoning Hierarchy before GRPO (ablation)Qwen2.5-VL-7B-Instruct · MindPower Benchmark actionSR: 0.28 -> 8.50Improved
add Mind-Reward GRPO after SFT (ablation)SFT model · MindPower Benchmark actionSR: 8.50 -> 11.75Improved
add Mind-Reward GRPO after SFT (ablation)SFT model · MindPower Benchmark actionAC: 10.48 -> 15.40Improved
add Mind-Reward GRPO after SFT (ablation)SFT model · MindPower Benchmark decisionSentence Transformer: 43.84 -> 47.12Improved
add Mind-Reward GRPO after SFT (ablation)SFT model · MindPower Benchmark perceptionSentence Transformer: 52.72 -> 59.93Improved
add Mind-Reward GRPO directly to the base model without SFT (ablation)Qwen2.5-VL-7B-Instruct · MindPower Benchmark decisionSentence Transformer: 26.56 -> 24.68No clear effect
add MindPower Reasoning Hierarchy prompting instead of direct decision and action output (ablation)GPT-4o · MindPower Benchmark actionaccuracy: 0.82 -> 2.91Improved
add MindPower Reasoning Hierarchy instead of standard step-by-step reasoning (ablation)GPT-4o · MindPower Benchmark actionaccuracy: 0.90 -> 2.91Improved
add MindPower Reasoning Hierarchy instead of standard step-by-step reasoning (ablation)GPT-4o · MindPower Benchmark decisionaccuracy: 4.89% drop without itImproved
add MindPower Reasoning Hierarchy instead of direct output (ablation)GPT-4o · MindPower Benchmark decisionaccuracy: 1.24% drop without itImproved
fine-tune Qwen2.5-VL-7B with SFT plus Mind-RewardQwen2.5-VL-7B-Instruct (video) · MindPower Benchmark perceptionSentence Transformer: +20.04%Improved
fine-tune Qwen2.5-VL-7B with SFT plus Mind-RewardQwen2.5-VL-7B-Instruct · MindPower Benchmark decisionSentence Transformer: +23.33%Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy