English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The advent of Chat-GPT has led to a surge of interest in Embodied AI.However, many existing Embodied AI models heavily rely on massive interactionswith training environments, which may not be practical in real-worldsituations. To this end, the Maniskill2 has introduced a full-physicssimulation benchmark for manipulating various 3D objects. This benchmarkenables agents to be trained using diverse datasets of demonstrations andevaluates their ability to generalize to unseen scenarios in testingenvironments. In this paper, we propose a novel two-stage fine-tuning strategythat aims to further enhance the generalization capability of our model basedon the Maniskill2 benchmark. Through extensive experiments, we demonstrate theeffectiveness of our approach by achieving the 1st prize in all three tracks ofthe ManiSkill2 Challenge. Our findings highlight the potential of our method toimprove the generalization abilities of Embodied AI models and pave the way fortheir ractical applications in real-world scenarios. All codes and models ofour solution is available at https://github.com/xtli12/GXU-LIPE.git

A Two-stage Fine-tuning Strategy for Generalizable Manipulation Skill of Embodied AI