Research map / Reinforcement learning / Visual and representation RL
Intention-Conditioned Flow Occupancy Models
Chongyi Zheng, Seohong Park, Sergey Levine, Benjamin Eysenbach
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 implicit GPI (upper expectile) instead of standard GPI over finite intentions (ablation) | InFOM · quadruped jump and scene task 1 | return / success rate: 44% more performant, 8x smaller variance | Improved |
| add the variational intention encoder (vs FOM with one-step policy improvement) (ablation) | InFOM · quadruped jump and scene task 1 | return / success rate | Improved |
| use variational intention encoder instead of HILP or FB representations for FOM conditioning (ablation) | flow occupancy models · OGBench cube double task 1 and others | downstream performance: better on 3 of 4 tasks | Improved |
| pre-train InFOM and fine-tune vs alternative pre-training methods | InFOM · OGBench state-based manipulation | success rate: +36% over best baseline | Improved |
| InFOM on image inputs | InFOM · OGBench visual tasks | success rate: +31% over best baseline | Improved |
| InFOM vs baselines on easier ExORL domains | InFOM · ExORL cheetah and quadruped | return | No clear effect |
Closest papers by meaning
- Imagine Beyond ! Distributionally Robust Autoencoding for State Space Coverage in Online Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Stitching Sub-trajectories with Conditional Diffusion Model for Goal-Conditioned Offline RL (AAAI 2024 · similarity 0.86)
- Flow Matching for Offline Reinforcement Learning with Discrete Actions (NeurIPS 2026 · similarity 0.86)
- RLP: Reinforcement as a Pretraining Objective (ICLR 2026 · similarity 0.85)
- Flattening Hierarchies with Policy Bootstrapping (NeurIPS 2025 · similarity 0.85)
- Designing Reinforcement Learning for Diffusion Models: A Unified Path-Space View (NeurIPS 2026 · similarity 0.85)
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.
