Research map / Vision-language and multimodal / Robot action and navigation
Planning with an Embodied Learnable Memory
Priyam Parashar, Jacob Krantz, Matthew Chang, Kavit Shah, Xavier Puig, Roozbeh Mottaghi
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 DDAFT RL fine-tuning (ablation) | PP planner · PARTNR (GT perception) | success rate: 0.51 -> 0.66 | Improved |
| add DDAFT RL fine-tuning (ablation) | HD planner · PARTNR (GT perception) | success rate: 0.63 -> 0.68 | Improved |
| add DDAFT RL fine-tuning (ablation) | PP planner · PARTNR (learned perception) | success rate: 0.46 -> 0.58 | Improved |
| add DDAFT RL fine-tuning (ablation) | HD planner · PARTNR (learned perception) | success rate: 0.55 -> 0.58 | Improved |
| fine-tune 8B planner on human-demonstration traces instead of prompting 70B (ablation) | Llama · PARTNR (GT perception) | success rate: 0.51 -> 0.63 | Improved |
| fine-tune on human-demonstration traces instead of prompting (ablation) | Llama · PARTNR (learned perception) | success rate: 0.46 -> 0.55 | Improved |
| use EPM-based planner instead of DynaMem | DynaMem · PARTNR (learned perception) | success rate: 0.03 -> 0.58 | Improved |
| use DynaMem with ground-truth perception | DynaMem · PARTNR | success rate: 0.03 -> 0.17 | No clear effect |
| use DDAFT after HD, learned perception (ablation) | HD · PARTNR | planning cycles: 30.2 -> 23.6 | Improved |
| use HD planner in place of PP (ablation) | PP · PARTNR (learned perception) | extraneous actions: 0.28 -> 0.17 | Improved |
Closest papers by meaning
- Context-Aware Planning and Environment-Aware Memory for Instruction Following Embodied Agents (ICCV 2023 · similarity 0.90)
- Instruction-Augmented Long-Horizon Planning: Embedding Grounding Mechanisms in Embodied Mobile Manipulation (AAAI 2025 · similarity 0.90)
- World-aware Planning Narratives Enhance Large Vision-Language Model Planner (NeurIPS 2025 · similarity 0.89)
- EVLP: Learning Unified Embodied Vision-Language Planner with Reinforced Supervised Fine-Tuning (ICLR 2026 · similarity 0.88)
- Memo: Training Memory-Efficient Embodied Agents with Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Exploratory Retrieval-Augmented Planning For Continual Embodied Instruction Following (NeurIPS 2024 · similarity 0.88)
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.
