Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Context-Aware Planning and Environment-Aware Memory for Instruction Following Embodied Agents
Byeonghwi Kim, Jinyeon Kim, Yuyeong Kim, Cheolhong Min, Jonghyun Choi
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 context-aware planning (CAP) (ablation) | CAPEAM · ALFRED test unseen | SR: 42.25 -> 46.11 | Improved |
| add environment-aware memory (EAM) (ablation) | CAPEAM · ALFRED test unseen | SR: 41.66 -> 46.11 | Improved |
| add environment-aware memory (EAM) (ablation) | CAPEAM · ALFRED test seen | SR: 46.44 -> 51.79 | Improved |
| add both CAP and EAM versus neither (ablation) | CAPEAM · ALFRED test unseen | SR: 39.83 -> 46.11 | Improved |
| increase planner parameters from 164.95M to 651.01M without context (ablation) | LSTM planner without CAP · ALFRED validation | planned action accuracy: 74.18 -> 74.91 (unseen); 80.12 -> 79.27 (seen) | No clear effect |
| add context-aware planning versus larger model (ablation) | planner · ALFRED validation unseen | planned action accuracy: 74.91 -> 77.34 | Improved |
| use CAPEAM* with templated actions versus Prompter on seen environments | CAPEAM* · ALFRED test seen (templated) | SR: 49.38 -> 46.64 | Hurt |
| use CAPEAM versus prior methods without templated actions | CAPEAM · ALFRED test unseen | SR: 36.07 -> 43.69 | Improved |
Closest papers by meaning
- Exploratory Retrieval-Augmented Planning For Continual Embodied Instruction Following (NeurIPS 2024 · similarity 0.89)
- Embodied Agents Meet Personalization: Investigating Challenges and Solutions Through the Lens of Memory Utilization (ICLR 2026 · similarity 0.89)
- ThinkBot: Embodied Instruction Following with Thought Chain Reasoning (ICLR 2025 · similarity 0.88)
- Reasoning in Visual Navigation of End-to-end Trained Agents: A Dynamical Systems Approach (CVPR 2025 · similarity 0.88)
- World-aware Planning Narratives Enhance Large Vision-Language Model Planner (NeurIPS 2025 · similarity 0.87)
- EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents (ICML 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.
