Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Learning on the Go: A Meta-learning Object Navigation Model
Xiaorong Qin, Xinhang Song, Sixian Zhang, Xinyao Yu, Xinmiao Zhang, Shuqiang Jiang
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 test-time adaptation to the central distribution (LOG) (ablation) | DAT · iTHOR cross-environment (ALL) | Succ. (%): 78.54 -> 84.02 | Improved |
| add test-time adaptation to the central distribution (LOG) (ablation) | DAT · iTHOR cross-environment (ALL) | SPL (%): 44.12 -> 46.95 | Improved |
| add test-time adaptation to the central distribution (LOG) (ablation) | PEANUT · HM3D (test-standard) | Succ. (%): 60.2 -> 64.3 | Improved |
| add test-time adaptation to the central distribution (LOG) (ablation) | PEANUT · MP3D (val) | Succ. (%): 37.5 -> 40.3 | Improved |
| use meta-learning instead of supervised training before the same test-time finetuning (ablation) | DAT · iTHOR cross-environment (ALL) | Succ. (%): 79.32 -> 84.02 | Improved |
| use meta-learning instead of supervised training before the same test-time finetuning (ablation) | PEANUT · HM3D (test-standard) | Succ. (%): 61.3 -> 64.3 | Improved |
| use meta-learning instead of supervised training before the same test-time finetuning (ablation) | PEANUT · MP3D (val) | SPL (%): 14.5 -> 15.4 | Mixed |
| replace SAVN meta-learning with LOG | DAT · iTHOR cross-environment (ALL) | Succ. (%): 79.67 -> 84.02 | Improved |
| add SAVN meta-learning to DAT | DAT · iTHOR cross-environment (ALL) | Succ. (%): 82.02 -> 79.67 | Hurt |
| add LOG to PEANUT | PEANUT · MP3D to Gibson cross-simulator (ALL) | Succ. (%): 48.7 -> 63.9 | Improved |
| add LOG to DAT for repeated tasks in one environment | DAT · iTHOR single-environment adaptivity (ALL) | Succ. (%): 83.59 -> 86.43 | Improved |
| increase number of inner-loop gradient updates (ablation) | LOG · iTHOR/HM3D | accuracy | Mixed |
Closest papers by meaning
- Learning to Learn How to Learn: Self-Adaptive Visual Navigation Using Meta-Learning (CVPR 2019 · similarity 0.91)
- Auxiliary Tasks and Exploration Enable ObjectGoal Navigation (ICCV 2021 · similarity 0.90)
- Reasoning in Visual Navigation of End-to-end Trained Agents: A Dynamical Systems Approach (CVPR 2025 · similarity 0.89)
- Trajectory Diffusion for ObjectGoal Navigation (NeurIPS 2024 · similarity 0.89)
- C-NAV: Towards Self-Evolving Continual Object Navigation in Open World (NeurIPS 2025 · similarity 0.89)
- Navigation World Models (CVPR 2025 · 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.
