AI Paper Atlas

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

HUGE-Bench: A Benchmark for High-Level UAV Vision-Language-Action Tasks

Jingyu Guo, Ziye Chen, Ziwen Li, Zhengqing Gao, Jiaxin Huang, Hanlue Zhang, Fengming Huang, Yu Yao, Tongliang Liu, Mingming Gong

ECCV 2026 · Vision-language and multimodal › Robot action and navigation · approach: Vision-language-action foundation models · 9 citations (OpenAlex)

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
add implicit memory to OpenVLA (MemoryVLA)OpenVLA · HUGE-Bench test (high-level instructions)Avg. TCR: 0.112 -> 0.234Improved
add implicit memory to OpenVLA (MemoryVLA)OpenVLA · HUGE-Bench test (high-level instructions)NTP: 0.122 -> 0.549Improved
add depth maps as extra visual input to pi_0.5pi_0.5 · HUGE-Bench test (high-level instructions)CR: 0.018 -> 0.013Improved
add depth maps as extra visual input to pi_0.5pi_0.5 · HUGE-Bench test (high-level instructions)nDTW: 0.467 -> 0.459Hurt
add depth maps as extra visual input to pi_0.5pi_0.5 · HUGE-Bench test (high-level instructions)NTP: 0.618 -> 0.617No clear effect
use high-level instead of low-level instructionspi_0.5 · HUGE-Bench test (high-level instructions)Avg. TCR: 0.698 -> 0.581Hurt
use high-level instead of low-level instructionsFastVLM · HUGE-Bench test (high-level instructions)Avg. TCR: 0.476 -> 0.285Hurt
use high-level instead of low-level instructionspi_0 · HUGE-Bench test (high-level instructions)nDTW: 0.561 -> 0.443Hurt
use high-level instead of low-level instructionsOpenVLA · HUGE-Bench test (high-level instructions)CR: 0.008 -> 0.003Mixed
use pi_0.5 instead of pi_0pi_0 policy · HUGE-Bench test (high-level instructions)Avg. TCR: 0.558 -> 0.581Improved
use pi_0.5 instead of pi_0pi_0 policy · HUGE-Bench test (high-level instructions)CR: 0.011 -> 0.018Hurt
add explicit frame-history grids from past 20 or 50 meters to OpenVLA-style baseline (ablation)OpenVLA/MemoryVLA · HUGE-Benchtask performanceHurt

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