AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

3DWay: Generalizing Robot Manipulation via 3D Consistent Waypoints

Huang Ziqin, Yingyue Li, Chenyangguang Zhang, Ruida Zhang, Yuxin Chen, Gu Wang, Xingyu Liu, Masayoshi TOMIZUKA, Xiangyang Ji

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 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
use 3DWay-TD direct execution instead of pi0.5manipulation policy · RLBench (10 tasks, unseen)success rate (%): 18.8 -> 64.0Improved
use 3DWay-TD instead of pi0.5manipulation policy · VLABenchavg success rate (%): 31.3 -> 37.7Improved
3DWay-TD vs pi0.5 on semantic-instruction dimensionmanipulation policy · VLABench semantic-instructionsuccess rate (%): 25.8 -> 24.3Hurt
add 3DWay waypoints (zero-shot generator) to pi0 fine-tuningpi0 · RLBench 5 tasks, 10 demos/tasksuccess rate (%): 17.6 -> 38.1Improved
use few-shot fine-tuned waypoint generator instead of zero-shotpi0 + 3DWay · RLBench 5 tasks, 10 demos/tasksuccess rate (%): 38.1 -> 46.1Improved
add 3DWay augmentation with 10 demos vs standard with 50 demospi0 · RLBench 5 taskssuccess rate (%): 46.1 (10 demos) vs 46.7 (50 demos)Improved
add 3DWay to pi0 on real-world basic taskspi0 · Real world basic tasks (AgileX PIPER)avg success rate (%): 21.7 -> 65.8Improved
add multi-view consistent supervision instead of independent-view (ablation)3DWay generator · RLBench (unseen)success rate (%): 22.0 -> 64.0Improved
scale backbone from NVILA-2B to 15B (ablation)3DWay generator · RLBench (unseen)success rate (%): 50.1 -> 64.0Improved
perturb camera pose at test time (ablation)3DWay generator · RLBench (unseen)success rate (%): 64.0 -> 61.6 (small), 58.4 (large)Hurt
change robot embodiment and camera setup (ablation)3DWay-TD · Real worldavg success rate (%): 43.9 -> 45.0 / 45.5 / 44.7No clear effect

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