2 citations · 10 across the 39 of their papers we have counts for
7 papers · 1 filter
Prioritized Rollouts for Efficient World Model-based Vision-Language-Action Policy Optimization
Yifei Sheng, Haoxiang Ren, Zhilong Zhang +8
Vision-Language-Action (VLA) models have emerged as a powerful paradigm for embodied intelligence, but fine-tuning them with reinforcement learning (RL) remains constrained by the…
Scaling GUI Agents with Visual State Transitions
Xiangyan Liu, Kaixin Li, Haonan Wang +6
We introduce State Transition Pretraining (STP) as a new scaling axis for GUI agents. During the STP stage, we continually pretrain a unified multimodal model on visual state trans…
Predictive Divergence Masks for LLM RL
Xiangxin Zhou, Jiarui Yao, Penghui Qi +4
Reinforcement learning for large language models (LLMs) typically relies on trust-region masks to stabilize off-policy updates. The dominant PPO-style approach uses the sampled-tok…
Anticipation-VLA: Solving Long-Horizon Embodied Tasks via Anticipation-based Subgoal Generation
Zhilong Zhang, Wenyu Luo, Haonan Wang +9
Vision-Language-Action (VLA) models have emerged as a powerful paradigm for embodied intelligence, enabling robots to perform tasks based on natural language instructions and curre…
Visual Generation in the New Era: An Evolution from Atomic Mapping to Agentic World Modeling
Keming Wu, Zuhao Yang, Kaichen Zhang +24
Recent visual generation models have made major progress in photorealism, typography, instruction following, and interactive editing, yet they still struggle with spatial reasoning…
Towards Practical World Model-based Reinforcement Learning for Vision-Language-Action Models
Zhilong Zhang, Haoxiang Ren, Yihao Sun +6
Vision-Language-Action (VLA) models show strong generalization for robotic control, but finetuning them with reinforcement learning (RL) is constrained by the high cost and safety…
Transformers Are Born Biased: Structural Inductive Biases at Random Initialization and Their Practical Consequences
Siquan Li, Yao Tong, Haonan Wang +1
Transformers underpin modern large language models (LLMs) and are commonly assumed to be behaviorally unstructured at random initialization, with all meaningful preferences emergin…