Research map / Reinforcement learning / Visual and representation RL
Goal-Driven Reward by Video Diffusion Models for Reinforcement Learning
Qi Wang, Mian Wu, Yuyang Zhang, Mingqi Yuan, Wenyao Zhang, Haoxiang You, Yunbo Wang, Xin Jin, Xiaokang Yang, Wenjun Zeng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use GenReward video-diffusion goal reward instead of the environment dense reward | DreamerV3 · Meta-World Pick Out of Hole | episode return: 193 -> 582 | Improved |
| use GenReward instead of the environment dense reward | DreamerV3 · Meta-World Bin Picking | episode return: 398 -> 822 | Improved |
| use GenReward instead of the environment dense reward | DreamerV3 · Meta-World Shelf Place | episode return: 154 -> 814 | Improved |
| add video-level reward (video encoder similarity to generated goal video) (ablation) | GenReward · Meta-World Pick Place | episode return | Improved |
| add frame-level forward-backward reward with CLIP-selected goal frame (ablation) | GenReward · Meta-World Pick Place | episode return | Improved |
| increase video-level reward weight alpha too far or too low (ablation) | GenReward · Meta-World Pick Place | episode return | Mixed |
| use goal videos generated from RT-1 or RLBench instead of Bridge (ablation) | GenReward · Meta-World Pick Place | episode return | Improved |
Closest papers by meaning
- Learning to Reach Goals via Diffusion (ICML 2024 · similarity 0.89)
- Designing Reinforcement Learning for Diffusion Models: A Unified Path-Space View (NeurIPS 2026 · similarity 0.88)
- Task-Agnostic Pre-training and Task-Guided Fine-tuning for Versatile Diffusion Planner (ICML 2025 · similarity 0.88)
- ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understanding (NeurIPS 2025 · similarity 0.88)
- Text-Aware Diffusion for Policy Learning (NeurIPS 2024 · similarity 0.88)
- Subtask-Aware Visual Reward Learning from Segmented Demonstrations (ICLR 2025 · similarity 0.87)
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.
