6 citations · 12 across the 13 of their papers we have counts for
14 papers
Neighbor GRPO: Contrastive ODE Policy Optimization Aligns Flow Models
Dailan He, Guanlin Feng, Xingtong Ge +6
Group Relative Policy Optimization (GRPO) has shown promise in aligning image and video generative models with human preferences. However, applying it to modern flow matching model…
One Model for All Tasks: Leveraging Efficient World Models in Multi-Task Planning
Yuan Pu, Yazhe Niu, Jia Tang +3
In heterogeneous multi-task decision-making, tasks not only exhibit diverse observation and action spaces but also vary substantially in their underlying complexities. While conven…
SafeWork-R1: Coevolving Safety and Intelligence under the AI-45 Law
Shanghai AI Lab, :, Yicheng Bao +115
We introduce SafeWork-R1, a cutting-edge multimodal reasoning model that demonstrates the coevolution of capabilities and safety. It is developed by our proposed SafeLadder framewo…
CleanS2S: Single-file Framework for Proactive Speech-to-Speech Interaction
Yudong Lu, Yazhe Niu, Shuai Hu +1
CleanS2S is a framework for human-like speech-to-speech interaction that advances conversational AI through single-file implementation and proactive dialogue capabilities. Our syst…
Empowering LLMs in Decision Games through Algorithmic Data Synthesis
Haolin Wang, Xueyan Li, Yazhe Niu +2
Large Language Models (LLMs) have exhibited impressive capabilities across numerous domains, yet they often struggle with complex reasoning and decision-making tasks. Decision-maki…
Revisiting Generative Policies: A Simpler Reinforcement Learning Algorithmic Perspective
Jinouwen Zhang, Rongkun Xue, Yazhe Niu +4
Generative models, particularly diffusion models, have achieved remarkable success in density estimation for multimodal data, drawing significant interest from the reinforcement le…
Pretrained Reversible Generation as Unsupervised Visual Representation Learning
Rongkun Xue, Jinouwen Zhang, Yazhe Niu +4
Recent generative models based on score matching and flow matching have significantly advanced generation tasks, but their potential in discriminative tasks remains underexplored.…
PsyDI: Towards a Personalized and Progressively In-depth Chatbot for Psychological Measurements
Xueyan Li, Xinyan Chen, Yazhe Niu +2
In the field of psychology, traditional assessment methods, such as standardized scales, are frequently critiqued for their static nature, lack of personalization, and reduced part…
AMEX: Android Multi-annotation Expo Dataset for Mobile GUI Agents
Yuxiang Chai, Siyuan Huang, Yazhe Niu +5
AI agents have drawn increasing attention mostly on their ability to perceive environments, understand tasks, and autonomously achieve goals. To advance research on AI agents in mo…
UniZero: Generalized and Efficient Planning with Scalable Latent World Models
Yuan Pu, Yazhe Niu, Zhenjie Yang +3
Learning predictive world models is crucial for enhancing the planning capabilities of reinforcement learning (RL) agents. Recently, MuZero-style algorithms, leveraging the value e…
ReZero: Boosting MCTS-based Algorithms by Backward-view and Entire-buffer Reanalyze
Chunyu Xuan, Yazhe Niu, Yuan Pu +3
Monte Carlo Tree Search (MCTS)-based algorithms, such as MuZero and its derivatives, have achieved widespread success in various decision-making domains. These algorithms employ th…
A Perspective of Q-value Estimation on Offline-to-Online Reinforcement Learning
Yinmin Zhang, Jie Liu, Chuming Li +4
Offline-to-online Reinforcement Learning (O2O RL) aims to improve the performance of offline pretrained policy using only a few online samples. Built on offline RL algorithms, most…
LightZero: A Unified Benchmark for Monte Carlo Tree Search in General Sequential Decision Scenarios
Yazhe Niu, Yuan Pu, Zhenjie Yang +6
Building agents based on tree-search planning capabilities with learned models has achieved remarkable success in classic decision-making problems, such as Go and Atari. However, i…
ACE: Cooperative Multi-agent Q-learning with Bidirectional Action-Dependency
Chuming Li, Jie Liu, Yinmin Zhang +5
Multi-agent reinforcement learning (MARL) suffers from the non-stationarity problem, which is the ever-changing targets at every iteration when multiple agents update their policie…