Research map / Vision-language and multimodal / Robot action and navigation
Hypothesis Graph Refinement: Hypothesis-Driven Exploration with Cascade Error Correction for Embodied Navigation
Peixin Chen, Guoxi Zhang, Jianwei Ma, Qing Li
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 hypothesis module (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SR: 67.71 -> 72.41 | Improved |
| add semantic hypothesis module (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SPL: 50.02 -> 56.22 | Improved |
| add semantic hypothesis module (ablation) | HGR · A-EQA | LLM-Match: 48.4 -> 55.9 | Improved |
| add semantic hypothesis module (ablation) | HGR · EM-EQA | LLM-Match: 50.5 -> 58.3 | Improved |
| add cascade correction (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SR: 68.61 -> 72.41 | Improved |
| add cascade correction (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SPL: 51.12 -> 56.22 | Improved |
| add cascade correction (ablation) | HGR · A-EQA | LLM-Match: 49.6 -> 55.9 | Improved |
| cascade correction instead of local delete only (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SR: 70.85 -> 72.41 | Improved |
| add both semantic hypothesis and cascade correction (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SR: 63.42 -> 72.41 | Improved |
| refutation threshold 0.3 instead of 0.5 (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SR: 72.41 -> 66.2 | Hurt |
| refutation threshold 0.7 instead of 0.5 (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SR: 72.41 -> 67.8 | Hurt |
| use GPT-4o instead of heuristic co-occurrence predictor (ablation) | HGR · GOAT-Bench (subset, 278 subtasks) | SPL: 52.13 -> 56.22 | Improved |
Closest papers by meaning
- Hierarchical 3D Scene Graph Construction and Belief-based Planning for Semantic Navigation (ECCV 2026 · similarity 0.91)
- Expand Your SCOPE: Semantic Cognition over Potential-Based Exploration for Embodied Visual Navigation (AAAI 2026 · similarity 0.90)
- Toward a Causally-Inspired Evolving World Model for Vision-and-Language Navigation in Continuous Environments (TPAMI 2026 · similarity 0.90)
- NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models (AAAI 2024 · similarity 0.88)
- Hierarchical Semantic-Augmented Navigation: Optimal Transport and Graph-Driven Reasoning for Vision-Language Navigation (NeurIPS 2025 · similarity 0.88)
- 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.
