Research map / Vision-language and multimodal / Cross-modal learning
Semantic Audio-Visual Navigation in Continuous Environments
Yichen Zeng, Hebaixu Wang, Meng Liu, Yu Zhou, Chen Gao, Kehan Chen, Gongping Huang
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 memory-augmented goal descriptor network (vs SAVi) | audio-visual navigation agent · SAVN-CE clean | SR (%): 25.6 -> 37.7 | Improved |
| add goal descriptor network (GDN) (ablation) | MAGNet · SAVN-CE clean | SR (%): 32.4 -> 37.7 | Improved |
| add episodic memory to GDN (ablation) | MAGNet · SAVN-CE clean | SR (%): 33.9 -> 37.7 | Improved |
| add self-motion cues to GDN (ablation) | MAGNet · SAVN-CE clean | SR (%): 34.3 -> 37.7 | Improved |
| add episodic memory to GDN (ablation) | MAGNet · SAVN-CE clean | SWS (%): 8.9 -> 10.6 | Improved |
| add GDN without memory or self-motion (ablation) | MAGNet · SAVN-CE clean | SR (%): 32.4 -> 33.9 | Improved |
| use MAGNet in distracted environments vs SAVi | navigation agent · SAVN-CE distracted | SR (%): 18.5 -> 19.3 | Improved |
| use MAGNet in distracted environments vs SAVi | navigation agent · SAVN-CE distracted | DSR (%): 6.4 -> 7.8 | Hurt |
| give oracle goal labels after sound ceases (Oracle2 vs Oracle1) | oracle variant · SAVN-CE clean | SR (%): 41.4 -> 75.0 | Improved |
| MAGNet GDN vs SAVi GDN during sounding period | GDN · SAVN-CE clean | F<=20 deg: 0.148 -> 0.290 | Improved |
| add distractor sounds | all methods · SAVN-CE | SR (%): 37.7 -> 19.3 (MAGNet) | Hurt |
Closest papers by meaning
- Towards Audio-Visual Navigation in Noisy Environments: A Large-Scale Benchmark Dataset and an Architecture Considering Multiple Sound-Sources (AAAI 2025 · similarity 0.90)
- RILA: Reflective and Imaginative Language Agent for Zero-Shot Semantic Audio-Visual Navigation (CVPR 2024 · similarity 0.90)
- RAO-Nav: Probing Omni-Language Models for Zero-shot Semantic Audio-Visual Navigation (NeurIPS 2026 · similarity 0.90)
- CAVEN: An Embodied Conversational Agent for Efficient Audio-Visual Navigation in Noisy Environments (AAAI 2024 · similarity 0.90)
- SAVVY: Spatial Awareness via Audio-Visual LLMs through Seeing and Hearing (NeurIPS 2025 · similarity 0.88)
- Hear, Localize, and Reason: Spatially Aware Scene Understanding for Audio-visual LLMs (NeurIPS 2026 · 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.
