11 papers
Optimal Skill Selection for LLM Agents with Provable Bicriteria Guarantees
Yu Chen, Ruishuo Chen, Xun Wang +2
Loading reusable skill documents into a bounded context window has become a primary way large language model (LLM) agents acquire task-specific capabilities, which makes skill sele…
MLREF: Efficient Module Reuse for Reward Design in Reinforcement Learning via Large Language Models
Chenglin Liu, Xun Wang, Ruishuo Chen +2
Reward function design remains a bottleneck in reinforcement learning. While large language models (LLMs) have enabled automated reward generation, existing methods generate and re…
When Context Returns: Toward Robust Internalization in On-Policy Distillation
Xun Wang, Ruishuo Chen, Zhuoran Li +2
Recent work has shown that on-policy distillation can internalize privileged context, such as system prompts or task hints, into a student model so that the context is no longer ne…
Reparameterization Flow Policy Optimization
Hai Zhong, Zhuoran Li, Xun Wang +1
Reparameterization Policy Gradient (RPG) has emerged as a powerful paradigm for model-based reinforcement learning, enabling high sample efficiency by backpropagating gradients thr…
Diffusing to Coordinate: Efficient Online Multi-Agent Diffusion Policies
Zhuoran Li, Hai Zhong, Xun Wang +3
Online Multi-Agent Reinforcement Learning (MARL) is a prominent framework for efficient agent coordination. Crucially, enhancing policy expressiveness is pivotal for achieving supe…
From Solo to Symphony: Orchestrating Multi-Agent Collaboration with Single-Agent Demos
Xun Wang, Zhuoran Li, Yanshan Lin +2
Training a team of agents from scratch in multi-agent reinforcement learning (MARL) is highly inefficient, much like asking beginners to play a symphony together without first prac…
OM2P: Offline Multi-Agent Mean-Flow Policy
Zhuoran Li, Xun Wang, Hai Zhong +3
Generative models, especially diffusion and flow-based models, have been promising in offline multi-agent reinforcement learning. However, integrating powerful generative models in…
Reparameterization Proximal Policy Optimization
Hai Zhong, Xun Wang, Zhuoran Li +1
By leveraging differentiable dynamics, Reparameterization Policy Gradient (RPG) achieves high sample efficiency. However, current approaches are hindered by two critical limitation…
Beyond the Proxy: Trajectory-Distilled Guidance for Offline GFlowNet Training
Ruishuo Chen, Xun Wang, Rui Hu +2
Generative Flow Networks (GFlowNets) excel at sampling diverse, high-reward objects. In many practical applications where active reward queries are infeasible, these models must be…
Beyond Shallow Behavior: Task-Efficient Value-Based Multi-Task Offline MARL via Skill Discovery
Xun Wang, Zhuoran Li, Hai Zhong +1
As a data-driven approach, offline MARL learns superior policies solely from offline datasets, ideal for domains rich in historical data but with high interaction costs and risks.…
Offline-to-Online Multi-Agent Reinforcement Learning with Offline Value Function Memory and Sequential Exploration
Hai Zhong, Xun Wang, Zhuoran Li +1
Offline-to-Online Reinforcement Learning has emerged as a powerful paradigm, leveraging offline data for initialization and online fine-tuning to enhance both sample efficiency and…