35 citations · 64 across the 28 of their papers we have counts for
13 papers · 1 filter
Segmental Advantage Estimation: Enhancing PPO for Long-Context LLM Training
Xue Gong, Qi Yi, Ziyuan Nan +8
Training Large Language Models (LLMs) for reasoning tasks is increasingly driven by Reinforcement Learning with Verifiable Rewards (RLVR), where Proximal Policy Optimization (PPO)…
Efficient Diffusion Planning with Temporal Diffusion
Jiaming Guo, Rui Zhang, Zerun Li +7
Diffusion planning is a promising method for learning high-performance policies from offline data. To avoid the impact of discrepancies between planning and reality on performance,…
QiMeng-Attention: SOTA Attention Operator is generated by SOTA Attention Algorithm
Qirui Zhou, Shaohui Peng, Weiqiang Xiong +11
The attention operator remains a critical performance bottleneck in large language models (LLMs), particularly for long-context scenarios. While FlashAttention is the most widely u…
QiMeng-TensorOp: Automatically Generating High-Performance Tensor Operators with Hardware Primitives
Xuzhi Zhang, Shaohui Peng, Qirui Zhou +12
Computation-intensive tensor operators constitute over 90\% of the computations in Large Language Models (LLMs) and Deep Neural Networks.Automatically and efficiently generating hi…
Context Shift Reduction for Offline Meta-Reinforcement Learning
Yunkai Gao, Rui Zhang, Jiaming Guo +10
Offline meta-reinforcement learning (OMRL) utilizes pre-collected offline datasets to enhance the agent's generalization ability on unseen tasks. However, the context shift problem…
Efficient Symbolic Policy Learning with Differentiable Symbolic Expression
Jiaming Guo, Rui Zhang, Shaohui Peng +8
Deep reinforcement learning (DRL) has led to a wide range of advances in sequential decision-making tasks. However, the complexity of neural network policies makes it difficult to…
Contrastive Modules with Temporal Attention for Multi-Task Reinforcement Learning
Siming Lan, Rui Zhang, Qi Yi +10
In the field of multi-task reinforcement learning, the modular principle, which involves specializing functionalities into different modules and combining them appropriately, has b…
Online Prototype Alignment for Few-shot Policy Transfer
Qi Yi, Rui Zhang, Shaohui Peng +10
Domain adaptation in reinforcement learning (RL) mainly deals with the changes of observation when transferring the policy to a new environment. Many traditional approaches of doma…
Conceptual Reinforcement Learning for Language-Conditioned Tasks
Shaohui Peng, Xing Hu, Rui Zhang +7
Despite the broad application of deep reinforcement learning (RL), transferring and adapting the policy to unseen but similar environments is still a significant challenge. Recentl…
Object-Category Aware Reinforcement Learning
Qi Yi, Rui Zhang, Shaohui Peng +6
Object-oriented reinforcement learning (OORL) is a promising way to improve the sample efficiency and generalization ability over standard RL. Recent works that try to solve OORL t…
Causality-driven Hierarchical Structure Discovery for Reinforcement Learning
Shaohui Peng, Xing Hu, Rui Zhang +9
Hierarchical reinforcement learning (HRL) effectively improves agents' exploration efficiency on tasks with sparse reward, with the guide of high-quality hierarchical structures (e…
Eden: A Unified Environment Framework for Booming Reinforcement Learning Algorithms
Ruizhi Chen, Xiaoyu Wu, Yansong Pan +12
With AlphaGo defeats top human players, reinforcement learning(RL) algorithms have gradually become the code-base of building stronger artificial intelligence(AI). The RL algorithm…
Hindsight Value Function for Variance Reduction in Stochastic Dynamic Environment
Jiaming Guo, Rui Zhang, Xishan Zhang +6
Policy gradient methods are appealing in deep reinforcement learning but suffer from high variance of gradient estimate. To reduce the variance, the state value function is applied…