Kangrui Wang

ROSETTA: Constructing Code-Based Reward from Unconstrained Language Preference

Spatial Mental Modeling from Limited Views

Internalizing World Models via Self-Play Finetuning for Agentic RL

SENTINEL: A Multi-Level Formal Framework for Safety Evaluation of LLM-based Embodied Agents

ERA: Transforming VLMs into Embodied Agents via Embodied Prior Learning and Online Reinforcement Learning

EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents

VAGEN: Reinforcing World Model Reasoning for Multi-Turn VLM Agents