Research map / Reinforcement learning / Visual and representation RL
Semantic Temporal Abstraction via Vision-Language Model Guidance for Efficient Reinforcement Learning
Tian-Shuo Liu, Xu-Hui Liu, Ruifeng Chen, Lixuan Jin, Pengyuan Wang, Zhilong Zhang, Yang Yu
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes a method that uses vision-language model guidance to extract semantically meaningful temporally extended skills for more efficient reinforcement learning under sparse rewards.
Problem
Temporal abstraction for RL: unsupervised skill discovery often lacks semantic information and yields erroneous or scattered skill extraction, hurting sparse-reward RL efficiency.
Approach
Uses vision-language model guidance to produce semantic temporal abstractions (skills) for RL, rather than purely unsupervised skill discovery.
Closest papers by meaning
- TTRV: Test-Time Reinforcement Learning for Vision Language Models (CVPR 2026 · similarity 0.89)
- Dynamic Contrastive Skill Learning with State-Transition Based Skill Clustering and Dynamic Length Adjustment (ICLR 2025 · similarity 0.89)
- VLMs-Guided Representation Distillation for Efficient Vision-Based Reinforcement Learning (CVPR 2025 · similarity 0.89)
- Goal-Oriented Skill Abstraction for Offline Multi-Task Reinforcement Learning (ICML 2025 · similarity 0.89)
- Code as Reward: Empowering Reinforcement Learning with VLMs (ICML 2024 · similarity 0.89)
- GoalLadder: Incremental Goal Discovery with Vision-Language Models (NeurIPS 2025 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
