Research map / Vision-language and multimodal / Robot action and navigation
Collaborative Instance Object Navigation: Leveraging Uncertainty-Awareness to Minimize Human-Agent Dialogues
Francesco Taioli, Edoardo Zorzi, Gianni Franchi, Alberto Castellini, Alessandro Farinelli, Marco Cristani, Yiming Wang
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 the Self-Questioner and Skip-Question together to AIUTA (ablation) | AIUTA · CoIN ablation split | SR: 9.21 -> 14.47 | Improved |
| add only the Self-Questioner (no Skip-Question) (ablation) | AIUTA · CoIN ablation split | SR: 9.21 -> 9.87 | Improved |
| add only Skip-Question (no Self-Questioner) (ablation) | AIUTA · CoIN ablation split | SR: 9.21 -> 8.55 | Hurt |
| add Skip-Question when Self-Questioner is on (ablation) | AIUTA · CoIN ablation split | NQ: 4.6 -> 1.68 | Improved |
| add user interaction via AIUTA on top of zero-shot VLFM | VLFM · CoIN-Bench Val Seen Synonyms | SR: 0.00 -> 14.38 | Improved |
| use AIUTA instead of trained baselines on unseen categories | AIUTA · CoIN-Bench Val Unseen | SR: 2.61 -> 6.67 (vs OVON) | Improved |
| training-free AIUTA versus OVON trained on GOAT-Bench on seen-synonym categories | AIUTA · CoIN-Bench Val Seen Synonyms | SR: 15.88 -> 14.38 | Hurt |
| replace real human with VLM-simulated user (ablation) | AIUTA · CoIN-Bench subset (40 episodes) | SR: 42.50 -> 42.50 | No clear effect |
| use Normalized Entropy instead of Energy Score for VLM uncertainty (ablation) | AIUTA Self-Questioner · IDKVQA | Effective Reliability Phi_c=1: 20.45 -> 21.12 | Improved |
| use Normalized Entropy instead of MaxProb for uncertainty (ablation) | AIUTA Self-Questioner · IDKVQA | Effective Reliability Phi_c=1: 15.94 -> 21.12 | Improved |
| use Normalized Entropy threshold instead of Energy Score threshold under data scarcity | uncertainty threshold tau · IDKVQA subsampled | normalized Phi_c=1 sensitivity | Improved |
Closest papers by meaning
- AwareVLN: Reasoning with Self-awareness for Vision-Language Navigation (CVPR 2026 · similarity 0.89)
- Human-Aware Vision-and-Language Navigation: Bridging Simulation to Reality with Dynamic Human Interactions (NeurIPS 2024 · similarity 0.88)
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.87)
- BOP-ASK: Object-Interaction Reasoning for Vision-Language Models (CVPR 2026 · similarity 0.87)
- NavBench: Probing Multimodal Large Language Models for Embodied Navigation (NeurIPS 2025 · similarity 0.87)
- Personalized Instance-based Navigation Toward User-Specific Objects in Realistic Environments (NeurIPS 2024 · 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.
