← Back to all news
arXiv Machine Learning By Yatai Ji, Zhengqiu Zhu, Yong Zhao, Yue Hu, Fanglong Yao, Chen Gao, Pengfei Zhu, Quanjun Yin

SearchWorld: Spatial Value-Grounded Imagination for UAV Object Search via World Models

Read the original on arXiv Machine Learning →

SearchWorld introduces a recurrent state‑space world model that couples explicit spatial memory with value‑guided imagination for UAV object search. The model maintains bird’s‑eye‑view exploration and obstacle memory, decodes a task‑aware spatial value layer, and uses a cognition‑action network to improve policy through imagined rollouts without a separate critic. Experiments on UAV‑ON show a 23.8% success rate, surpassing the best published agent (19.5%) and achieving 35.5% oracle success while remaining robust on unseen scenes.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Computer Vision
6d ago

Learning to Think and Explore in 3D World

The paper introduces ACT3D, a vision‑language model that learns to think and explore in 3D environments using only single‑image inputs and action‑based interactions. By fine‑tuning with trajectory supervision and then applying reinforcement learning driven by final‑answer correctness, ACT3D develops an exploration policy that adapts to the question and improves accuracy across six benchmarks, outperforming both proprietary and open‑source baselines while using fewer input frames. The study shows that outcome‑based RL can teach VLMs to decide when and how to acquire visual evidence, leading to better reasoning and spatial understanding.

By Yijiang Li, Bingyang Wang, Nuno Vasconcelos
llmsreinforcement-learningfine-tuningmultimodalbenchmarks
More like this →
arXiv AI
Aug 19

Towards Unified World Models for Visual Navigation via Memory-Augmented Planning and Foresight

The paper introduces UniWM, a unified, memory‑augmented world model that merges egocentric visual foresight and planning into a single multimodal autoregressive backbone. By grounding action selection in visually imagined outcomes and using a hierarchical memory to fuse short‑term perception with long‑term trajectory context, UniWM aligns prediction with control and improves navigation stability. Experiments on four challenging benchmarks and the 1X Humanoid Dataset show up to 30% higher success rates, reduced trajectory errors, zero‑shot generalization to unseen datasets, and scalability to high‑dimensional humanoid navigation.

By Yifei Dong, Fengyi Wu, Guangyu Chen, Lingdong Kong, Qiyu Hu, Yuxuan Zhou, Xu Zhu, Jingdong Sun, Jun-Yan He, Qi Dai, Alexander G. Hauptmann, Zhi-Qi Cheng
agentsroboticsmultimodalbenchmarks
More like this →