14 papers
Ego2Robot: Scalable Robot Data Synthesis from Egocentric Human Data
Ye Wang, Pei Lin, Xiong-Hui Chen +12
Learning generalizable robot manipulation policies requires large-scale and diverse demonstration data. Egocentric human manipulation videos offer rich scene and task diversity, an…
Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models
Haoqi Yuan, Zhixuan Liang, Anzhe Chen +20
Foundation models in language and multimodality achieve strong generalization by aligning heterogeneous data under a unified formulation and training at scale. In this report, we i…
RedEdit: Agentic Red-Teaming of Image Safety Classifiers via MCTS-Guided Photo-Editing
Weilin Lin, Ziqi Lin, Zhenxing Zhou +4
Image safety classifiers serve as a critical component of contemporary content moderation systems on the internet. However, their resilience against user-style malicious image edit…
OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents
Rui Yang, Qianhui Wu, Yuxi Chen +7
Building capable visual web agents requires long-horizon reasoning, precise grounding, and robust interaction with dynamic real-world websites. Despite rapid progress, the stronges…
AsyncWebRL: Efficient Asynchronous Reinforcement Learning for Multi-Step Visual Web Agents
Hao Bai, Rui Yang, Chenlu Ye +3
Training vision-language web agents with multi-step RL is compute-intensive, with two dominant forms of inefficiency: idle GPUs in synchronous RL, and trajectories that use more st…
Qwen-VLA: Unifying Vision-Language-Action Modeling across Tasks, Environments, and Robot Embodiments
Qiuyue Wang, Mingsheng Li, Jian Guan +37
Embodied intelligence is often studied through specialized models for individual tasks such as manipulation or navigation, resulting in fragmented capabilities and limited generali…
PRO-CUA: Process-Reward Optimization for Computer Use Agents
Yifei He, Rui Yang, Hao Bai +2
Computer use agents (CUAs) have shown strong potential for automating complex digital workflows, yet their training remains constrained by costly live environment interaction and l…
Skill-Aligned Annotation for Reliable Evaluation in Text-to-Image Generation
Abdelrahman Eldesokey, Merey Ramazanova, Ahmad Sait +4
Text-to-image (T2I) generation has advanced rapidly, making reliable evaluation critical as performance differences between models narrow. Existing evaluation practices typically a…
Orchard: An Open-Source Agentic Modeling Framework
Baolin Peng, Wenlin Yao, Qianhui Wu +11
Agentic modeling aims to transform LLMs into autonomous agents capable of solving complex tasks through planning, reasoning, tool use, and multi-turn interaction with external envi…
Recursive Multi-Agent Systems
Jiaru Zou, Rui Pan, Ruizhong Qiu +8
Recursive or looped language models have recently emerged as a new scaling axis by iteratively refining the same model computation over latent states to deepen reasoning. We extend…
GUI-Libra: Training Native GUI Agents to Reason and Act with Action-aware Supervision and Partially Verifiable RL
Rui Yang, Qianhui Wu, Zhaoyang Wang +8
Open-source native GUI agents still lag behind closed-source systems on long-horizon navigation tasks. This gap stems from two limitations: a shortage of high-quality, action-align…
Chain-of-Experts: Unlocking the Communication Power of Mixture-of-Experts Models
Zihan Wang, Rui Pan, Jiarui Yao +7
We propose Chain-of-Experts (CoE), a new Mixture-of-Experts (MoE) architecture that introduces sequential expert communication within each layer. Unlike traditional MoE models, whe…
MiCRo: Mixture Modeling and Context-aware Routing for Personalized Preference Learning
Jingyan Shen, Jiarui Yao, Rui Yang +5
Reward modeling is a key step in building safe foundation models when applying reinforcement learning from human feedback (RLHF) to align Large Language Models (LLMs). However, rew…
MergeBench: A Benchmark for Merging Domain-Specialized LLMs
Yifei He, Siqi Zeng, Yuzheng Hu +3
Model merging provides a scalable alternative to multi-task training by combining specialized finetuned models through parameter arithmetic, enabling efficient deployment without t…