Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Learning to Map for Active Semantic Goal Navigation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 active training rather than offline training | L2M · Occupancy prediction | Acc (%): 65.2 -> 68.1 | Helped |
| Use active training rather than offline training | L2M · Semantic prediction | IoU (%): 20.1 -> 25.6 | Helped |
| Train using the proposed information-gain objective instead of BALD | L2M-UpperBound · Object-goal navigation | Success (%): 33.4 ± 1.0 -> 36.3 ± 1.0 | Helped |
| Train using the proposed information-gain objective instead of predictive entropy | L2M-UpperBound · Object-goal navigation | Success (%): 31.3 ± 1.0 -> 36.3 ± 1.0 | Helped |
| Replace the learned stop decision with an oracle stop decision (ablation) | L2M-Active-UpperBound · Hard object-goal navigation episodes | Success | Helped |
| Replace the local policy with simulator shortest paths (ablation) | L2M-Active-UpperBound · Hard object-goal navigation episodes | Success | Helped |
| Replace both the learned stop decision and local policy with oracle stop and simulator shortest paths (ablation) | L2M-Active-UpperBound · Hard object-goal navigation episodes | Success: over 80% | Helped |
Closest papers by meaning
- Distilling LLM Prior to Flow Model for Generalizable Agent’s Imagination in Object Goal Navigation (NeurIPS 2025 · similarity 0.91)
- Imagine Before Go: Self-Supervised Generative Map for Object Goal Navigation (CVPR 2024 · similarity 0.90)
- Hierarchical Semantic-Augmented Navigation: Optimal Transport and Graph-Driven Reasoning for Vision-Language Navigation (NeurIPS 2025 · similarity 0.89)
- LangMap: A Human-Verified Benchmark for Hierarchical Open-Vocabulary Goal Navigation (NeurIPS 2026 · similarity 0.89)
- BeliefMapNav: 3D Voxel-Based Belief Map for Zero-Shot Object Navigation (NeurIPS 2025 · similarity 0.88)
- Learning with a Mole: Transferable latent spatial representations for navigation without reconstruction (ICLR 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.