Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Auxiliary Tasks and Exploration Enable ObjectGoal Navigation
Joel Ye, Dhruv Batra, Abhishek Das, Erik Wijmans
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 Semantic Goal Exists (SGE) input feature (ablation) | 4-Action agent · MP3D VAL | success %: 20.3 -> 34.4 | Improved |
| add Semantic Goal Exists (SGE) input feature (ablation) | 4-Action agent · MP3D VAL | SPL %: 4.14 -> 9.58 | Improved |
| add Coverage Prediction auxiliary task (ablation) | 4-Action agent · MP3D VAL | success %: 32.4 -> 34.4 | Improved |
| add Generalized Inverse Dynamics auxiliary task (ablation) | 4-Action agent · MP3D VAL | success %: 33.7 -> 34.4 | Improved |
| add Action Distribution Prediction auxiliary task (ablation) | 4-Action agent · MP3D VAL | success %: 34.1 -> 34.4 | No clear effect |
| use tethered policy instead of direct sparse-reward finetuning (ablation) | 6-Action agent · MP3D VAL | success %: 23.1 -> 26.6 | Improved |
| use tethered policy instead of direct sparse-reward finetuning (ablation) | 6-Action agent · MP3D VAL | SPL %: 8.43 -> 9.79 | Improved |
| add tethered sparse-reward policy | 6-Action agent · MP3D VAL | SPL %: SPL 7.60 -> 9.79, success 30.8 -> 26.6 | Mixed |
| finetune with RedNet segmentation | 4-Action agent · MP3D VAL | success %: 34.4 -> 33.1 | Hurt |
| finetune with RedNet segmentation | 6-Action agent · MP3D VAL | success %: 30.8 -> 34.6 | Improved |
| add LOOK_UP and LOOK_DOWN actions (6-action vs 4-action) | agent with GT segmentation · MP3D VAL subset (300 episodes, GT segm.) | success %: 43.3 -> 58.0 | Improved |
| replace GT segmentation with RedNet predicted segmentation | 6-Action agent · MP3D VAL subset (300 episodes) | success %: 58.0 -> 30.8 | Hurt |
Closest papers by meaning
- PONI: Potential Functions for ObjectGoal Navigation With Interaction-Free Learning (CVPR 2022 · similarity 0.90)
- 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)
- EfficientNav: Towards On-Device Object-Goal Navigation with Navigation Map Caching and Retrieval (NeurIPS 2025 · similarity 0.88)
- Seeing through Uncertainty: Robust Task-Oriented Optimization in Visual Navigation (NeurIPS 2025 · similarity 0.87)
- Reasoning in Visual Navigation of End-to-end Trained Agents: A Dynamical Systems Approach (CVPR 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.
