MineExplorer is a benchmark designed to assess the open‑world exploration abilities of multimodal large language models (MLLMs) in Minecraft. It filters out tasks that rely heavily on Minecraft‑specific knowledge, organizes tasks into ReAct‑style capabilities, and composes atomic tasks into implicit multi‑hop challenges. A multi‑agent synthesis workflow creates reliable task graphs, sandbox scenes, and rule‑based milestone evaluators, and human evaluation confirms its superiority over a single‑agent baseline. Experiments show that while advanced MLLMs can handle many single‑hop tasks, they struggle with longer trajectories that require coordinating hidden prerequisites, and larger models or different thinking modes do not consistently improve performance.
By Tianjie Ju, Yueqing Sun, Zheng Wu, Wei Zhang, Yaqi Huo, Xi Su, Qi Gu, Xunliang Cai, Gongshen Liu, Zhuosheng Zhang
EmbodiedSkills is a unified framework that treats each skill decision as an execution proposal, checking prerequisites and verifying outcomes during long‑horizon vision‑language‑action tasks. It connects high‑level skill selection, bounded low‑level VLA execution, and post‑action verification through a fixed executable‑skill interface, enabling easy replacement of low‑level policies and recording of structured trajectories for supervision and adaptation. Instantiated with Qwen3‑VL and OpenPI/pi0.5 on RoboTwin 2.0 and LIBERO, the framework achieves high success rates (86.20% and 97.40% respectively) and demonstrates effective memory‑dependent task performance.
By Wei Wang, Wenqiao Zhang, Yutong Lin, Yuqian Yuan, Tianwei Lin, Jinhao Mao, Zhenxuan Fan, Mingjian Gao, Yang Dai, Wentong Li, Zheqi Lv, Zheng Dong, Yingjie Niu, Jiaqi Zhu, Jun Xiao, Chao Li, Yueting Zhuang
A central capability of embodied agents is to accomplish complex objectives through sequences of interdependent tasks. Yet existing visual goal-conditioned policies underlying these agents are typical...
arXiv:2607. 05377v1 Announce Type: cross Abstract: While recent Vision-Language-Action (VLA) models show promise toward generalist manipulation policies, they struggle with long-horizon tasks due to their Markovian nature-relying solely on current observations.
By Jiaqi Peng, Xiqian Yu, Delin Feng, Yuqiang Yang, Wenzhe Cai, Jing Xiong, Ganlin Yang, Jinliang Zheng, Jiafei Cao, Xueyuan Wei, Jiangmiao Pang, Yuan Shen, Tai Wang
arXiv:2606. 27251v1 Announce Type: cross Abstract: Building persistent embodied agents in unstructured environments demands unified orchestration of heterogeneous tools spanning both cyber (APIs, IoT) and physical (manipulation, navigation) domains, coupled with autonomous recovery from physical failures that inevitably arise over extended operation.
By Junhao Shi, Zezheng Huai, Siyin Wang, Jia Chen, Yubang Wang, Zhaoye Fei, Hechang Chen, Jingjing Gong, Xipeng Qiu, Yu-Gang Jiang
The paper introduces Instruct-to-Act, a system that decouples planning and control by combining a vision‑language model (VLM) planner with a world‑model controller. The VLM generates sparse, high‑level text instructions, while the controller executes them at high frequency, trained via relabeling rollouts with synthetic instructions and joint optimization of behavior cloning, reward, and world‑model objectives. Across seven embodied environments—including multi‑agent settings—this approach outperforms controller‑only and direct VLM action methods, maintains fast control, and allows swapping pretrained VLM planners without fine‑tuning, achieving competitive results with strong baselines on most tasks.
By Zineng Tang, Kelsey R. Allen, Sjoerd van Steenkiste, Ishita Dasgupta, Alane Suhr
Attacca is a new approach for training visual goal‑conditioned policies that can handle continuous long‑horizon task execution in embodied agents. It decouples goal images from the execution environment by sampling class‑compatible masked goals from other worlds, enabling the policy to ground targets in the current view via a target‑mask prediction head. The method also introduces behavioral‑phase conditioning to distinguish Search, Approach, and Interact stages, and achieves significant performance gains on Minecraft tasks, outperforming baselines by up to 7× on long‑horizon scenarios.
By Gyusik Seo, Jaehong Yoon
arXiv:2607. 10350v1 Announce Type: new Abstract: Recent VLM and VLA systems have improved robotic perception and action prediction, yet long-horizon embodied agents still require a general runtime layer for reasoning, memory, tool use, verification, and cross-embodiment execution.
By Jiayi Tian, Shiao Liu, Yuting Xu, Jia Lu, Zihao Guan, Honglin Han, Di Yang, Minqi Gu, Yifei Qian, Tianlin Zhang, Yanqing Zhu, Zeqian Ye, Menglin Yang, Fei Wang, Xu Hu, Xiuxian Li, Wei Zhang, Shihui Su, Yiyan Ji, Jingbo Wang, Ziteng Feng, Jiaheng Liu, Zhaoxiang Zhang, Xiaolong Wu, Mingyang Yin, Zedong Chu, Mu Xu
arXiv:2605.13119v2 Announce Type: replace-cross
Abstract: Vision-language-action (VLA) models are effective robot action executors, but they remain limited on long-horizon tasks due to the dual burde...
By Zixing Lei, Changxing Liu, Yichen Xiong, Minhao Xiong, Yuanzhuo Ding, Zhipeng Zhang, Weixin Li, Siheng Chen
The paper introduces Instruct-to-Act, a system that decouples high‑level planning from low‑latency control by combining a vision‑language model (VLM) planner with a world‑model controller. The VLM generates sparse, high‑level text instructions, while the controller executes them autonomously at high frequency. Experiments across seven embodied environments, including multi‑agent settings, show that this approach outperforms both controller‑only and direct VLM action‑generation methods, maintains fast control, and allows swapping in different pretrained VLM planners without fine‑tuning.
arXiv:2604. 03208v2 Announce Type: replace Abstract: World models are a promising path to zero-shot embodied control through planning.
By Wancong Zhang, Basile Terver, Artem Zholus, Soham Chitnis, Harsh Sutaria, Mido Assran, Randall Balestriero, Amir Bar, Adrien Bardes, Yann LeCun, Nicolas Ballas
The paper introduces 2AM, a system that separates memory and action execution in long‑horizon robot manipulation. 2AM stores task memory exclusively in a multimodal Agent, while a single RGB‑based, stateless Action Model performs motion based on language and optional 2D hints. On the LIBERO‑Mem benchmark, 2AM achieves 76.3% average completion—over 61 points higher than the best baseline—demonstrating that agent‑side memory and precise steering of the Action Model can substantially improve performance.
By Yutong Hu, Fengjiao Chen, Xuezhi Cao, Renaud Detry