AgentVLN: a 3B VLM brain tops R2R-CE and RxR-CE in real time on Jetson
jiqizhixin · x · 2026-09-19
AgentVLN (ECCV 2026) introduces a "VLM-as-Brain" paradigm for vision-and-language navigation: Qwen2.5-VL-3B handles high-level decisions while a modular skill library (SLAM/perception/navigation) executes actions — no heavy 3D vision modules needed.
- A QD-PCoT mechanism generates fine-grained exploration under occlusion, blind spots and trajectory drift, curbing long-horizon error accumulation without extra parameters
- Leads the mainstream R2R-CE and RxR-CE benchmarks while running in real time on Jetson edge devices
More from Embodied
- DISCO distilled tree search named CoRL 2026 best-paper finalist, top 6 of 687 — yining_hong · 2026-11-04
- Berkeley's Value Forcing: enforcing Bellman consistency at every noise level for accurate RL value functions — berkeley_ai · 2026-11-04
- 22 Creative SAM 3.1 Use Cases From Meta's AI Developer Hackathon, From Wildfire Detection to Document Reassembly — Vjeux · 2026-11-04
- Ex-OpenAI robotics intern breaks down viral Astra robot-painting demo on Cerebras podcast — AccBalanced · 2026-11-04
- Embodied Agent Arena: Closed-Weight VLMs Lead All Five Robot Task Domains, Open-Weight Models Lag — Haojian Huang · 2026-11-04
- PrimeBot's PrimeSim: a generalist world model simulator trained on cross-embodiment demos — Travis Davies · 2026-11-04