Shaohui Peng

Affiliations

Institute of Computing Technology, Chinese Academy of SciencesInstitute of Software, Chinese Academy of Sciences

papers

total citations

papers (11)

Context Shift Reduction for Offline Meta-Reinforcement Learning

NEURIPS 2023arXiv

citations

ANPL: Towards Natural Programming with Interactive Decomposition

NEURIPS 2023arXiv

citations

Contrastive Modules with Temporal Attention for Multi-Task Reinforcement Learning

NEURIPS 2023arXiv

citations

OCEAN-MBRL: Offline Conservative Exploration for Model-Based Offline Reinforcement Learning

AAAI 2024

citations

SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection

CVPR 2025

citations

Hypothesis, Verification, and Induction: Grounding Large Language Models with Self-Driven Skill Learning

AAAI 2024

citations

Decompose a Task into Generalizable Subtasks in Multi-Agent Reinforcement Learning

NEURIPS 2023

citations

Shaohui Peng

Affiliations

papers (11)

Context Shift Reduction for Offline Meta-Reinforcement Learning

ANPL: Towards Natural Programming with Interactive Decomposition

Contrastive Modules with Temporal Attention for Multi-Task Reinforcement Learning

Efficient Symbolic Policy Learning with Differentiable Symbolic Expression

Emergent Communication for Numerical Concepts Generalization

Prompt-based Visual Alignment for Zero-shot Policy Transfer

Emergent Communication for Rules Reasoning

OCEAN-MBRL: Offline Conservative Exploration for Model-Based Offline Reinforcement Learning

SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection

Hypothesis, Verification, and Induction: Grounding Large Language Models with Self-Driven Skill Learning

Decompose a Task into Generalizable Subtasks in Multi-Agent Reinforcement Learning

papers (11)

Context Shift Reduction for Offline Meta-Reinforcement Learning

ANPL: Towards Natural Programming with Interactive Decomposition

Contrastive Modules with Temporal Attention for Multi-Task Reinforcement Learning

Efficient Symbolic Policy Learning with Differentiable Symbolic Expression

Emergent Communication for Numerical Concepts Generalization

Prompt-based Visual Alignment for Zero-shot Policy Transfer

Emergent Communication for Rules Reasoning

OCEAN-MBRL: Offline Conservative Exploration for Model-Based Offline Reinforcement Learning

SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection

Hypothesis, Verification, and Induction: Grounding Large Language Models with Self-Driven Skill Learning

Decompose a Task into Generalizable Subtasks in Multi-Agent Reinforcement Learning