AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

AURA: Multi-modal Shared Autonomy for Urban Navigation

Yukai Ma, Honglin He, Selina Song, Wayne Wu, Bolei Zhou

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Embodied multimodal LLM agents

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 changedWhereMeasuredEffect
finetune InternVL3-2B on the synthesized MM-CoS instruction labels (ablation)InternVL3-2B · Instruction following on MM-CoSROUGE-L: 0.167 -> 0.532Improved
finetuned InternVL3-2B versus larger non-finetuned InternVL3-8B (ablation)InternVL3-2B · Instruction following on MM-CoSIntent Score: 2.818 -> 4.885Improved
add SIE together with visual prompts (ablation)AURA · Instruction following on MM-CoSROUGE-L: 0.532 -> 0.581Improved
add SIE together with visual prompts (ablation)AURA · Instruction following on MM-CoSIntent Score: 4.885 -> 5.446Improved
use SIE alone without visual prompt versus finetuned VLM with visual prompt (ablation)AURA · Instruction following on MM-CoSROUGE-L: 0.532 -> 0.534No clear effect
remove semantic (visual prompt) encoding from instruction input (ablation)AURA end-to-end planning · drafting and arrowing planning, 2s L2L2 distance at 2sHurt
add geometry encoding to visual prompts (ablation)AURA end-to-end planning · long-horizon instruction lagL2 distance at 2sImproved
add geometry encoding to visual prompts (ablation)AURA end-to-end planning · short-horizon instruction lagL2 distance at 2sNo clear effect
use arrowing instruction instead of retrained CityWalker point-goal baselineAURA · MM-CoS open-loopL2 error at 2s: 0.786 -> 0.473Improved
use drafting instruction rather than other instruction typesAURA · MM-CoS open-loopmAP: 0.750 -> 0.844Improved
give humans high-level instructions instead of point-goal guidanceAURA shared control · pseudo-simulationHuman Operation Frequency: 0.96 -> 0.498Improved
use AURA instead of baseline navigation models in real-world sidewalkshared autonomy · Real-world pilot, 2.8 kmHuman Operation Ratio (%): 9.74 -> 1.73Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy