AI Paper Atlas

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

RoboAnnotatorX: A Comprehensive and Universal Annotation Framework for Accurate Understanding of Long-horizon Robot Demonstration

Longxin Kou, Fei Ni, Yan Zheng, Peilong Han, Jinyi Liu, Haiqin Cui, Rui Liu, Jianye Hao

ICCV 2025 · Vision-language and multimodal › Robot action and navigation · approach: Multimodal prompt-driven manipulation · 0 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 video-level token stream (ablation)RoboAnnotatorX · RoboX-VQAtemporal localizationImproved
add clip-level token stream on top of scene tokens (ablation)RoboAnnotatorX · RoboX-VQAtemporal reasoningImproved
add scene-level tokens for semantic understanding (ablation)RoboAnnotatorX · RoboX-VQAscene understandingImproved
add simulated data to training (ablation)RoboAnnotatorX · RoboX-VQAaccuracyImproved
add QA-based instruction tuning (ablation)RoboAnnotatorX · RoboX-VQAaccuracyImproved
use curriculum three-stage training instead of mixed training (ablation)RoboAnnotatorX · RoboX-VQAaccuracyImproved
fine-tune LLaMA-VID on RoboX-VQALLaMA-VID · RoboX-VQAAvg IoU: 13.43 -> 25.38Improved
multi-scale encoder versus fine-tuned LLaMA-VIDRoboAnnotatorX · RoboX-VQAAvg IoU: 25.38 -> 69.29Improved
fine-tune LLaMA-VID on RoboX-VQALLaMA-VID · RoboX-VQA OOD embodiment-levelaccuracy (%): 12.51 -> 35.73Improved
multi-scale encoder versus fine-tuned LLaMA-VIDRoboAnnotatorX · RoboX-VQA OOD scene-levelaccuracy (%): 30.03 -> 58.32Improved
add training-free foundation models to MLLM (NILS, MVU)general MLLM · RoboX-VQAAvg IoU: 2.01 and 17.47 vs 42.94 for GPT-4oNo clear effect
use RoboAnnotatorX annotations for planningOpenVLA · LohoRavensplan accuracy (%): 65.4% -> 83.5%Improved

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