# LAMP: Lift Image-Editing as General 3D Priors for Open-world Manipulation

Source: [LAMP: Lift Image-Editing as General 3D Priors for Open-world Manipulation](<https://www.sota2.com/research/paper/69d884407d0172140ec77c70>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-23T05:00:12.895Z

Authors: Jingjing Wang, Zhengdong Hong, Chong Bao, Yuke Zhu, Junhan Sun, Guofeng Zhang

Published: 2026-04-09T17:14:00.000Z

## Abstract

Human-like generalization in open-world remains a fundamental challenge for robotic manipulation. Existing learning-based methods, including reinforcement learning, imitation learning, and vision-language-action-models (VLAs), often struggle with novel tasks and unseen environments. Another promising direction is to explore generalizable representations that capture fine-grained spatial and geometric relations for open-world manipulation. While large-language-model (LLMs) and vision-language-model (VLMs) provide strong semantic reasoning based on language or annotated 2D representations, their limited 3D awareness restricts their applicability to fine-grained manipulation. To address this, we propose LAMP, which lifts image-editing as 3D priors to extract inter-object 3D transformations as continuous, geometry-aware representations. Our key insight is that image-editing inherently encodes rich 2D spatial cues, and lifting these implicit cues into 3D transformations provides fine-grained and accurate guidance for open-world manipulation. Extensive experiments demonstrate that \\codename delivers precise 3D transformations and achieves strong zero-shot generalization in open-world manipulation. Project page: https://zju3dv.github.io/LAMP/.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Lid Covering on Real-world manipulation](<https://www.sota2.com/research/sota/lid-covering-on-real-world-manipulation.md>): Lid covering; Real-world manipulation; Success Rate: 80.

- [Object-centric manipulation on Real-world 10 object-centric tasks (Success Rate)](<https://www.sota2.com/research/sota/object-centric-manipulation-on-real-world-10-object-centric-tasks-success-rate.md>): Object-centric manipulation; Real-world 10 object-centric tasks; Egg Placing Success Rate: 60.

- [Articulated-object manipulation on Real-world 3 articulated-object tasks](<https://www.sota2.com/research/sota/articulated-object-manipulation-on-real-world-3-articulated-object-tasks.md>): Articulated Object Manipulation; Real-world 3 articulated-object tasks; Drawer Opening Success Rate: 60.

- [Pencil Insertion on Real-world manipulation](<https://www.sota2.com/research/sota/pencil-insertion-on-real-world-manipulation.md>): Pencil Insertion; Real-world manipulation; Success Rate: 7.

- [Ring Stacking on Real-world manipulation](<https://www.sota2.com/research/sota/ring-stacking-on-real-world-manipulation.md>): Ring stacking; Real-world manipulation; Success Rate: 0.8.

- [Drawer Closing on Real-world robot manipulation](<https://www.sota2.com/research/sota/drawer-closing-on-real-world-robot-manipulation.md>): Drawer Closing; Real-world robot manipulation; Success Rate: 7.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.08475>)

[PDF](<https://arxiv.org/pdf/2604.08475.pdf>)