20 results for “Zero-Shot Object-Goal Navigation”
CS papers onlyHybrid search: Keyword + semantic, ranked by combined score.ⓘ
Want pure semantic search? Try claim verification →
Qi Chai, Wenhao Shen, Nanjie Yao, Yue Xia +4 more
This paper proposes a self-evolving framework for Zero-Shot Object-Goal Navigation that enables continuous test-time improvement.
Xinhao Liu, Yunsheng Tian, Yuheng Qiu, Pieter Abbeel +3 more
The paper improves object-goal navigation by explicitly grounding the goal, achieving state-of-the-art results on six benchmarks.
Daichi Azuma, Taiki Miyanishi, Koya Sakamoto, Shuhei Kurita +5 more
This paper proposes NavWAM, a diffusion-transformer policy that enables direct use of visual foresight for robot control in goal-conditioned visual navigation.
Jiazhao Zhang, Gengze Zhou, Hale Yin, Yiyang Huang +29 more
Qwen-RobotNav is a scalable navigation model that can be reconfigured for different tasks and environments.
Jisoo Kim, Sangwon Baik, Taeksoo Kim, Sungjoo Kim +3 more
The paper presents a zero-shot framework for long-horizon dexterous manipulation that grounds language instructions into executable 3D task plans from calibrated multi-view RGB images.
Zhenghao "Mark'' Peng, Honglin He, Quanyi Li, Yukai Ma +1 more
This paper proposes a VLM-Planner interface to improve sidewalk navigation by fusing a vision-language model with a learning-based planner.
Kangrui Wang, Linjie Li, Zhengyuan Yang, Shiqi Chen +6 more
The paper addresses the challenge of multi-turn view planning for VLMs by proposing an iterative framework that uses self-exploration and view graph distillation, significantly improving planning perf…
Tianle Zeng, Hanjing Ye, Jianwei Peng, Jingwen Yu +2 more
The paper proposes a memory-augmented, traversability-aware framework for outdoor VLN that maintains stable, goal-consistent guidance even when semantic cues are interrupted or unavailable.
T2Mo is a novel framework that generates controllable dynamic 3D object shapes by combining explicit 3D trajectories for spatial guidance with natural language text semantics.
The paper introduces a novel two-stage framework to achieve robust, category-agnostic object localization in-context (ICL) by optimizing attention and minimizing localization error using reinforcement…
This paper introduces a method for language-grounded robot group joining, enabling robots to predict socially compliant joining poses in real-time.
Mingjian Gao, Zhaocheng Li, Haoyang Huang, Wenqiao Zhang +6 more
This paper proposes EvolvingNav, a method for embodied agents to navigate in environments where targets may move while unobserved.
This paper improves instruction following in vision-language-action models through flatness-preserving optimization during finetuning.
This paper analyzes the observability and consistency properties of VINS with anchored feature representations and presents methods to improve estimator consistency.
Zhizheng Liu, Honglin He, Vivek Alumootil, Akshat Pandya +3 more
The authors propose SidewalkBench, a comprehensive benchmark for visual navigation on urban sidewalks.
Rufeng Chen, Yue Chang, Xiaqiang Tang, Hechang Chen +1 more
PSG-Nav addresses open-vocabulary navigation uncertainty by constructing a 3D Probabilistic Scene Graph and using Multiverse Decision Making to sample multiple possible world settings for robust, glob…
Lingyi Hong, Weiren Cai, Jiayu Chen, Xinyu Zhou +6 more
This paper presents SimpleVLN, an architecture-preserving framework for Vision-and-Language Navigation (VLN) that unlocks the potential of native multimodal large language models (MLLMs) without task-…