Research map / Reinforcement learning / Offline RL
Instructed Diffuser With Temporal Condition Guidance for Offline Reinforcement Learning
Jifeng Hu, Yanchao Sun, Sili Huang, Siyuan Guo, Hechang Chen, Li Shen, Lichao Sun, Yi Chang, Dacheng Tao
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 |
|---|---|---|---|
| add historical condition (vs FAR-RTG-TS) (ablation) | TCD · Gym-MuJoCo successful comparison | mean score: 84.98 -> 85.67 | Improved |
| add immediate condition to HC-RTG-TS (ablation) | TCD · Gym-MuJoCo successful comparison | mean score: 85.31 -> 85.67 | Improved |
| add historical and immediate conditions to RTG-only (ablation) | RTG-TS · Gym-MuJoCo successful comparison | mean score: 82.87 -> 85.67 | Improved |
| add historical and immediate conditions to RTG-only (ablation) | RTG-TS · Gym-MuJoCo general comparison | mean score: 66.45 -> 77.65 | Improved |
| add historical condition (vs immediate only) (ablation) | TCD · Gym-MuJoCo general comparison | mean score: 72.33 -> 77.65 | Improved |
| replace U-net with transformer backbone (TFD) (ablation) | Diffuser · Gym-MuJoCo successful comparison | mean score: 85.67 -> 43.37 | Hurt |
| use distributional Q value as condition (DQD) (ablation) | Diffuser · Gym-MuJoCo successful comparison | mean score: 85.67 -> 34.79 | Hurt |
| replace statistical reward with linear reward regression (RR-TCD) (ablation) | TCD · Gym-MuJoCo successful comparison | mean score: 85.67 -> 79.81 | Hurt |
| replace statistical reward with reward quantile regression (RQR-TCD) (ablation) | TCD · Gym-MuJoCo successful comparison | mean score: 85.67 -> 85.06 | No clear effect |
| add temporal conditions (TCD vs DD) (ablation) | DD · Maze2D | mean score: 30.71 -> 37.10 | Improved |
| add temporal conditions (TCD vs DD) (ablation) | DD · Maze2D dense umaze | normalized score: 83.23 -> 29.77 | Hurt |
| add temporal conditions (TCD vs DD) (ablation) | DD · Gym-MuJoCo random datasets | mean score: 53.32 -> 55.32 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MOReL — MOReL: Model-Based Offline Reinforcement Learning (NeurIPS 2020) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Simple Hierarchical Planning with Diffusion (ICLR 2024 · similarity 0.90)
- Prior-Guided Diffusion Planning for Offline Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Stitching Sub-trajectories with Conditional Diffusion Model for Goal-Conditioned Offline RL (AAAI 2024 · similarity 0.88)
- Structural Information-based Hierarchical Diffusion for Offline Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Adding Conditional Control to Diffusion Models with Reinforcement Learning (ICLR 2025 · similarity 0.88)
- State-Covering Trajectory Stitching for Diffusion Planners (NeurIPS 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.
