Research map / Multi-agent systems and games / Human-AI and goal-conditioned
CogNav: Cognitive Process Modeling for Object Goal Navigation with LLMs
Yihan Cao, Jiazhao Zhang, Zhinan Yu, Shuzhen Liu, Zheng Qin, Qin Zou, Bo Du, Kai Xu
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 |
|---|---|---|---|
| use CogNav instead of the best prior method | TriHelper · HM3D val | SR: 62.0 -> 72.5 | Improved |
| use CogNav instead of the best prior method | SG-NAV · MP3D val | SR: 40.2 -> 46.6 | Improved |
| use CogNav instead of the best prior method | SG-NAV · RoboTHOR val | SR: 47.5 -> 54.6 | Improved |
| use CogNav instead of the best prior method | SG-NAV · MP3D val | SPL: 16.0 -> 16.1 | No clear effect |
| add Contextual Search state (ablation) | CogNav (BS+OT) · HM3D val | SR: 55.2 -> 57.6 | Improved |
| add Candidate Verification state (ablation) | CogNav (BS+CS+OT) · HM3D val | SR: 57.6 -> 67.2 | Improved |
| add Candidate Verification state (ablation) | CogNav (BS+CS+OT+TC) · HM3D val | SPL: 33.6 -> 26.2 | Hurt |
| add Candidate Verification state (ablation) | CogNav (BS+CS+OT+TC) · HM3D val | SR: 68.8 -> 72.5 | Improved |
| add Target Confirmation state (ablation) | CogNav (BS+CS+OT+CV) · HM3D val | SR: 67.2 -> 72.5 | Improved |
| add edges to map-to-landmark prompt (ablation) | CogNav · HM3D val | SR: 63.3 -> 68.2 | Improved |
| add room type to map-to-landmark prompt (ablation) | CogNav · HM3D val | SR: 71.6 -> 72.5 | Improved |
| add spatial information to agent-to-landmark prompt (ablation) | CogNav · HM3D val | SR: 64.5 -> 72.5 | Improved |
Closest papers by meaning
- 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 0.88)
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.88)
- Trajectory Diffusion for ObjectGoal Navigation (NeurIPS 2024 · similarity 0.88)
- NavBench: Probing Multimodal Large Language Models for Embodied Navigation (NeurIPS 2025 · similarity 0.87)
- VELMA: Verbalization Embodiment of LLM Agents for Vision and Language Navigation in Street View (AAAI 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.
