Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks
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 a learned active-completion policy instead of random action selection | Active observation completion · SUN360 | MSE(x1000): 31.88 -> 23.16 | Helped |
| Use a learned active-completion policy instead of random action selection | Active observation completion · ModelNet seen classes | MSE(x1000): 3.46 -> 3.25 | Helped |
| Use a learned active-completion policy instead of random action selection | Active observation completion · ModelNet unseen classes | MSE(x1000): 6.22 -> 5.65 | Helped |
| Learn which views to acquire instead of repeatedly taking the largest allowable action | Active observation completion · SUN360 | MSE(x1000): 30.76 -> 23.16 | Helped |
| Learn which views to acquire instead of repeatedly taking the largest allowable action | Active observation completion · ModelNet seen classes | MSE(x1000): 3.44 -> 3.25 | Helped |
| Learn which views to acquire instead of repeatedly taking the largest allowable action | Active observation completion · ModelNet unseen classes | MSE(x1000): 6.16 -> 5.65 | Helped |
| Learn view selection instead of choosing the most salient reachable view | Active observation completion · SUN360 | MSE(x1000): 27.00 -> 23.16 | Helped |
| Learn view selection instead of choosing the most salient reachable view | Active observation completion · ModelNet seen classes | MSE(x1000): 3.47 -> 3.25 | Helped |
| Learn view selection instead of choosing the most salient reachable view | Active observation completion · ModelNet unseen classes | MSE(x1000): 6.35 -> 5.65 | Helped |
| Rely on sensed camera elevation as an exploration advantage | Learned active-completion policy · SUN360 and ModelNet | MSE(x1000) | No clear effect |
Closest papers by meaning
- DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning (ICLR 2026 · similarity 0.86)
- Selective Visual Representations Improve Convergence and Generalization for Embodied AI (ICLR 2024 · similarity 0.86)
- Reasoning in Visual Navigation of End-to-end Trained Agents: A Dynamical Systems Approach (CVPR 2025 · similarity 0.86)
- Mind the GAP: Glimpse-based Active Perception improves generalization and sample efficiency of visual reasoning (ICLR 2025 · similarity 0.86)
- Learning with a Mole: Transferable latent spatial representations for navigation without reconstruction (ICLR 2024 · similarity 0.86)
- Artemis: Structured Visual Reasoning for Perception Policy Learning (ICML 2026 · similarity 0.85)
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.